Prosjektønsker kan registreres fra 1. april 2025.
Velg hva du ønkser å vise prosjekt for.
Introduction:
[ Vis hele beskrivelsen ]
The emergence of Self-Sovereign Identity (SSI), also known as decentralized technology in healthcare, offers a transformative approach to Electronic Health Records (EHR) management, emphasizing patient autonomy and data security. Similar to the rapid adoption of ChatGPT in artificial intelligence, the integration of SSI in healthcare has gained significant interest, promising a more secure, private, and patient-focused EHR system.
Objective:
This project aims to develop an architecture and prototype for a decentralized Electronic Health Record (EHR) management system based on existing literature. The architecture and prototype will be designed to closely match the current EHR setting in Norway but will operate decentralizedly.
Methodology:
It will have the following components.
These findings are expected to contribute to the existing literature on SSI and Software Engineering. This research will provide knowledge about the state of practice, challenges, and opportunities in integrating tools like decentralized identities into the software development process.
[ Skjul beskrivelse ]
In the evolving landscape of healthcare technology, ensuring patient privacy and security in Electronic Health Records (EHR) is paramount. Self-Sovereign Identity (SSI) offers a promising solution to empower patients with control over their identities and health data.
This proposal aims to contribute valuable insights into leveraging Self-Sovereign Identity for empowering patient privacy and security in Electronic Health Records, ultimately fostering a more patient-centered approach to healthcare data management.
Objectives:
Expected Outcomes:
[More information here]
The power group in Refinitiv Research&Forecast has been for several years responsible for modelling regional wind power output forecasts across several regions in the world.
With the extreme increase in renewables in Europe it is becoming increasingly important to identify in advance possible weather conditions which might lead to wind power curtailment. Curtailment of power is a common phenomenon in wind farms. It simply means that the wind turbine is made to operate at a capacity lower than it should at a certain wind speed. This can be forced for safety reasons by the plant owners, or for grid congestion reasons by the system operators. Developing algorithms for anticipating such curtailment is of crucial importance both for the system operators and for the traders, since it will lead to more precise wind power forecasts, and as a consequence to better decisions. Ideally such algorithms should have as input both weather conditions, system operator forecast plans, and price forecasts.
[Collaboration between NTNU, Sintef Digital and TU Delft]
The purpose of this master's thesis is to explore and evaluate the effectiveness of data-driven and machine learning-based methods for synthetic data generation in the ATM (Air Traffic Management) domain. This research proposal is in collaboration with the SynthAIr European project, which aims to increase the level of automation of the ATM system by delivering novel AI methods for synthetic data generation in aviation.
The proposed research will focus on developing and evaluating various techniques for synthetic data generation, including generative adversarial networks (GANs), variational autoencoders (VAEs), diffusion models, and other state-of-the-art machine learning methods. The study will explore and evaluate the quality of the generated synthetic data and its impact on the performance of machine learning models for specific operational use cases in the ATM domain, including turnaround time prediction, flight delay prediction, and passenger flow prediction.
The research will explore the challenges associated with synthetic data generation in the ATM domain, such as mixed datatypes, missing values, and complex dependencies over long periods of time. The study will also analyze and measure the impact of synthetic data for increasing the level of efficiency, robustness, and resilience when adopting AI for increased automation of ATM systems.
The proposed research will contribute to advancing the understanding of how synthetic data generation methods can overcome the challenges of limited data availability and accelerate the adoption of AI in the ATM domain. The research will also provide insights into the effectiveness of different data generation techniques for specific operational use cases in the ATM domain.
Overall, this master thesis will be conducted as a part of a larger european research project integrating the student within a team of experienced researchers from SINTEF Digital and TUD (Technical Univeristy of Delft) and will contribute to advancing the use of artificial intelligence and machine learning in the ATM domain.
Main Supervisor: Massimiliano Ruocco (NTNU/Sintef Digital) [massimiliano.ruocco@ntnu.no]
Co-supervisors: Alexei Sharpanskykh (TUD), Alfredo Clemente (SINTEF Digital)
This project aims to develop game mechanics that will motivate users to socialize and be physically active using Augmented Reality. It is part of the BitPet project, which aims for commercialization. Developers in BitPet will provide technical support.
The project will involve a study of existing theory, game concepts, and technology, the design and development of a game concept (both front-end and back-end), and an evaluation of the concept involving real users.
The front end will be developed in Unity.
This project requires two students.
The goal of this project is to design and develop new game concepts for a game where an exercise bike is used as a game controller in addition to traditional game input through multiple buttons. In addition to input from buttons, the player should control the game by using her/his fit to move the pedals. The goal of the game is to have fun that can last over time and get physical exercise. The game should be implemented in Unity using an API provided for the exercise bike controller.
The goals of this project are:
Research existing exergames and games that could fit this purposeDesign and implement a prototype game Provide input on the API used for the exergame frameworkEvaluate the game through user experiments
This project is for a group of two students, and experience with Unity is required.
In this project, the goal is to develop new game concepts and technologies for exergames - games where the player performs physical exercise. There are several approaches to exergames, and the challenge is to find the balance between something that is fun to play and getting real physical exercise from playing the game.
The first phase of the project will consist of a theoretical study of exergames and mechanisms for using games as motivators. The second phase focuses on implementing a prototype using various technologies. In the third and final phase, the prototype will be evaluated and tested.
This project requires a group of two students.
The goal of this project is the design, implementation, and evaluation of a collaborative learning game in which the students beat the game together and learn at the same time. The game will have to balance engagement and learning to make it fun and educational. Another requirement is that it must be a multiplayer game in which all the students in a call can participate at the same time.
The project will involve studying research on game-based learning, designing and implementing a concept, and evaluating the concept with users.
The project requires a group of two students.
The objective of this project is to develop an automatic system for reading traffic announcements that can broadcast messages to the public through the "traffic announce" channel or VMA in Sweden. The system will use speech recognition and text-to-speech technology to automatically read out traffic announcements and other important messages to drivers and passengers, regardless of whether they are listening to the radio.
Dataset: Data will be provided by NRK.
Risk and Challenges: One of the primary risks associated with this project is the potential for misinterpretation of traffic announcements, which could lead to confusion and potentially dangerous situations.
Expected Outcome: The expected outcome of this project is an automatic system for reading traffic announcements that is accurate and fast to ensure the synthesized speech to be provided timely and reliable to the public.
Significance: The significance of this project lies in its potential to improve road safety and transportation management by ensuring that critical messages are delivered to all users in a timely and accessible manner. By reducing the reliance on drivers to listen to the radio for traffic announcements, the system can help to reduce distractions and improve overall driving safety.
Reference:Ao, J., Wang, R., Zhou, L., Wang, C., Ren, S., Wu, Y., Liu, S., Ko, T., Li, Q., Zhang, Y. and Wei, Z., 2021. Speecht5: Unified-modal encoder-decoder pre-training for spoken language processing. arXiv preprint arXiv:2110.07205.
Description in which company/unit the thesis will be placed: The Norwegian Research Center for AI Innovation (NorwAI).
Problem Description: Deaf people in Norway have collectively developed a way to communicate by means of signing. Signing involves using hands as well as expressions. There are about 20 000 people using the Norwegian Sign Language today. NRK offers programmes in which sign language is featured to make content accessible to deaf Norwegians. In the long term, we aim to create systems that can autonomously communicate via sign language in the form of a social robot. One step on the way concerns the understanding of the communication partner. We aim to transcribe the signed information to text. Using texts opens up a wide variety of existing machine learning and natural language processing tools.
Thesis Description: The thesis will conduct a literature review on automatic sign language detection. Note that sign languages in different parts of the world differ as spoken languages do. Still, we expect that we can learn from existing work on different sign languages. Candidate(s) will collect video footage with signing content from public sources. Subsequently, the candidate(s) will train a machine learning model to recognize and automatically transcribe signed expressions. The evaluation will be done with previously unknown footage.
Data Description: We are in talks with partners about possible data resources. Getting the video footage from NRK should be feasible.
Challenges (business and/or research): Computer Vision has seen many improvements since the advent of deep learning. Still, video footage represents a comparatively large data size and demand more time for computation. NTNU has a large GPU cluster which can aid. Still, the candidate(s) will have to familiarize themselves with the system.
Supervisor (NTNU): Benjamin Kille (benjamin.u.kille@ntnu.no)
Description in which company/unit the thesis will be placed: This master thesis will be carried out at the Norwegian Research Center for AI Innovation (NorwAI), NTNU.
Problem Description: Recent advances in graph-based learning approaches have demonstrated their effectiveness in modelling users' preferences and items' characteristics for Recommender Systems (RSs). Various graph embedding techniques and graph neural networks have been proposed and incorporated into the representation learning of RSs, using direct or multi-hop connections within graphs to enrich the representations of the user and item nodes. These methods further improve the recommendation performance. With tremendous amount of recommendation algorithms being proposed, one critical issue has attracted much attention from researchers in the community: there are no effective benchmarks for evaluation. It, consequently, leads to two major concerns, namely unreproducible evaluation and unfair comparison. Several recent studies show that the results of baselines reported in numerous publications over the past five years are suboptimal. With a careful setup, the baselines can outperform most of the newly proposed methods. Different from other domains, e.g., computer vision, where mature benchmarks are available to fairly evaluate the proposed approaches, benchmarking recommendation is more challenging in two aspects: (1) there exists a lot of datasets from different platforms in each application domain. (2) there are different data-processing strategies, data splitting methods, evaluation metrics and parameter settings, etc. Thus, this master thesis project aims to investigate evaluation rigorousness (i.e., reproducibility and fairness) in graph-based recommendation and develop a user-friendly public toolkit for rigorous evaluation.
Thesis Description: The overall goal of this project and master thesis is to benchmark graph-based recommendation for reproducible evaluation and fair comparison. To achieve this goal, the project is divided into the following tasks: (1) To conduct a systematic literature review on graph-based recommendation. The candidates are expected to summarize essential factors related to evaluation, including utilized datasets, data pre-processing strategies, comparison baselines, loss function designs, negative sampling strategies, data splitting methods, evaluation metrics, and parameter tuning strategies. Then, the candidates need to analyze the influence of different factors on recommendation performance. (2) Based on the investigation results from (1), the candidates will create benchmarks containing the standardized procedures to improve the reproducibility and fairness of evaluation, and provide the performance of selected well-tuned state-of-the-art algorithms. This includes a user-friendly public toolkit for rigorous evaluation.
Data Description: This project will use four real-world datasets covering different domains and sizes. The first one is the Adressa Dataset, which is a Norwegian news dataset containing about 113 million events in connection with anonymized users over a 90-day period from 1 January to 31 January 2017. This dataset has been pre-processed and stored in the json-style format. The size of the dataset is around 16G. It is publicly available with the license CC BY-NC-SA 4.0. MovieLens-1M dataset released by GroupLens (https://grouplens.org/datasets/movielens/). It describes 5-star rating and free-text tagging activity from MovieLens, a movie recommendation service. The dataset contains 100836 ratings and 3683 tag applications across 9742 movies created by 610 users between March 29, 1996 and September 24, 2018. The size of the dataset is around 1M. Last.FM dataset released by GroupLens contains social networking, tagging and music artist listening information from a set of 2K users from Last.fm online music system. Specifically, The dataset consists of 1892 users, 17,632 artists. The size of the dataset is around 2.6M. DBLP-Citation-Network dataset consists of bibliography data in computer science and relevant domains, extracted from DBLP, ACM, Microsoft Academic Graph and other sources. This project will leverage the last version of the data, V13, containing 5,354,309 papers and 48,227,950 papers. The dataset provides information such as papers’ title, abstract, author, venue, keywords and citation relations, and is publicly available for research purposes. The size of the dataset is around 2.4GB after depressed.
Challenges (business and/or research): Benchmarking graph-based recommendation faces many challenges: (1) experimental datasets are from various domains. Even for the same dataset, it may have different versions covering different durations; (2) different data-preprocessing strategies result in various graph types which require different recommendation approaches and parameter tunning methods; (3) various recommendation scenarios, e.g. rating prediction, top-k recommendation or online recommendation, may require different evaluation metrics; (4) most importantly, most papers do not report details on data processing and parameter tuning setting, leading to inconsistent results in reproduction by different researchers.
Supervisor (NTNU): Lemei Zhang (lemei.zhang@ntnu.no), Peng Liu (peng.liu@ntnu.no)
Problem Description: Today, we frequently encounter conversational agents/chatbots / virtual assistants when engaging with customer services. These AI systems facilitate information access. They provide a way to approach our information needs in the form of a conversation, as we would encounter with human-to-human interaction. Conversational agents can be available 24/7 and help the employees to focus on more difficult requests. The very unique language of economic sectors presents a challenge for conversational agents. Organizations use special nomenclature and sometimes their own terms or brands.
Thesis Description: We are interested in the financial sector in Norway. The thesis will conduct a literature survey on conversational agents in the financial sector focusing on Norway. We strongly encourage investigating systems that are publicly available. Subsequently, the candidate(s) will implement an information access system collecting public information from the Norwegian banking sector. This information can include websites, documents, or multi-media content. The candidate(s) will feed the information into a conversational agent and create a proof of concept. We are particularly interested in the question how to keep the system up to date.
Data Description: The data should be publicly available. The candidates will have to crawl the data themselves. We will assist with advice on how to efficiently crawl documents and websites.
Challenges (business and/or research): Working with language presents a few challenges. Many AI systems require numerical input. Thus, the text has to be encoded numerically to be useful. The candidate(s) will have to combine knowledge about Natural Language Processing and Conversational Systems.
Problem Description: In the era of Open Data, many parliaments publish transcriptions of sessions. These documents can help to advance public education about politics. Speeches reflect politicians' viewpoints on political matters. Social Sciences have invested a lot of effort into analyzing such data. AI, in particular Natural Language Processing, promises to automate the process. Specifically, recent advances in Deep Neural Networks, such as Transformers, have pushed the capabilities of AI systems. We seek to explore the use of AI for automated political speech analysis.
Thesis Description: The thesis will take existing corpora of transcribe parliament speeches and apply Natural Language Processing techniques. The necessary activities involve inspecting and transforming the data, designing an experiment to compare various AI techniques and reporting the findings. We would like to consider political speeches from parliaments with under-resourced languages, such as Norwegian, Swedish, Danish, or Finnish. This will help to provide better access to political information in Scandinavia. The AI part can either be formulated as a classification or clustering task. The former would attempt to predict the party affiliation of the speaker. The latter would seek to uncover commonalities among the speeches and possible find language patterns that reflect political viewpoints.
Data Description: As described earlier, most parliaments publish their proceedings digitally. The degree of pre-processing varies. in Norway, there is the Stortinget data set (https://www.nb.no/sprakbanken/en/resource-catalogue/oai-nb-no-sbr-58/). Germany's Bundestag's Proceedings are published online (https://opendiscourse.de/daten-und-methodik). The candidate(s) can select also data from different parliaments as long as their language skills allow them to evaluate the outcome.
Challenges (business and/or research): Working with text entails a set of challenges. Unlike numerical data, most machine learning methods are not immediately applicable. Instead, the data has to be transferred into a numerical representation.
Supervisor (NTNU): Benjamin Kille (benjamin.u.kille@ntnu.no) and Tu My Doan
Description in which company/unit the thesis will be placed: This master thesis will be carried out under the project "Generation of Large-Scale Norwegian GPT-2 Language Models" supported by the Norwegian Research Center for AI Innovation (NorwAI).
Problem Description: Today there are two official forms of written Norwegian, Bokmål and Nynorsk, each with its own variants. Even though the two written languages are considered equal, only 5-10% of Norwegian texts are written in Nynorsk. Nynorsk is therefore a low-resource language within a low-resource language. Translation between Bokmål and Nynorsk is not available in Google Translate, and all this research is very much in its infancy. On the technical side, almost existing neural machine translation (NMT) systems are based on the encoder-decoder framework and require a large sentence-aligned parallel corpus for model training. However, the lack of such data in Bokmål and Nynorsk renders the most commonly used NMT approaches ineffective. Meanwhile, Generative Pre-trained Transformer (GPT), which captures the rich contextual knowledge from large scale web texts, has gained tremendous success in a variety of natural language processing tasks. Thus, the upcoming question is whether the GPT language model can be used for machine translation with only limited parallel data.
Thesis Description: The overall goal of this project and master thesis is to develop an efficient GPT based neural machine translation model for Bokmål and Nynorsk. To achieve this goal, the project is divided into the following tasks: (1) Investigate the possible ways of utilizing pre-trained language models for translation tasks. As part of this, the candidates are expected to perform a state-of-the-art literature review and summarize the strength and weaknesses of existing approaches. (2) Based on the investigation results from (1), the candidates need to design an efficient approach to incorporating the GPT language model into neural machine translation model. This includes evaluating the method with respect to applicability.
Data Description: The Bokmål and Nynorsk parallel data come from the Norwegian Colossal Corpus at the National Library of Norway. This dataset has 200k translation units (TU) i.e. aligned pairs of sentences in the respective languages extracted from textbooks of various subjects and newspapers, giving a total size of 28.6 MB. It is publicly available with the license CC BY 4.0. The GPT language model is provided by the Norwegian Research Center for AI Innovation, which is trained from scratch for Norwegian and has 318.6M parameters.
Challenges (business and/or research): (1) With a small bilingual training corpus, it would be hard for the NMT model to learn how to recognize and translate named entities, which might sacrifice translation performance. (2) Most of the instances in the training corpus are short texts. How to deal with the translation of long texts is another challenge. (3) Fine-tuning is the prevalent paradigm for using large pre-trained language models to perform downstream tasks (e.g., translation and summarization), but it requires updating and storing all the parameters of the LM. Considering the size of current LMs, how to reduce training time and storage overhead when building and deploying machine translation systems is an urgent problem to be solved.
Problem Description: Over the past decade, there has been a trend toward leveraging and adapting approaches proposed by Natural Language Processing (NLP) research like Word2Vec, GRU, and Attention for recommender systems (RecSys). The phenomenon is especially noticeable for session-based and sequential recommendation where the sequential processing of users interactions is analogous to the language modeling task. More recently, Pre-trained Language Models (PLM) have shown the remarkable capability of understanding a variety of natural language tasks given task descriptions (e.g., prompts) and only a few or even zero demonstrations. With the significant increase in model size and pre-training data amount, PLMs encode extensive world knowledge and can even correctly answer more factoid questions than the models that explicitly use external knowledge bases. Moreover, recently-proposed prompt-based tuning approaches further improve the data efficiency — PLMs can be 100x more data-efficient when converting various natural language processing tasks such as text classification, question answering, natural language inference to cloze tasks with a prompt. This master thesis project aims to explore pre-trained language model based recommender systems, which use powerful PLMs as recommender systems by reformulating recommendation as a language modeling task. Specifically, there are two challenging problems to be tackled: (1) Can we use pre-trained language models for Cold-Start Recommendation? (2) Can we fine-tune pre-trained language models to improve the effectiveness and efficiency of recommendations?
Thesis Description: The overall goal of this project and master thesis is to explore the ability of pre-trained language models for session-based recommendation and alleviating the cold-start issues. To achieve this goal, the project is divided into the following tasks: (1) Investigate the possibility of applying existing pre-trained language models to solve the above challenging problems in session-based recommendation. As part of this, the candidates are expected to perform a state-of-the-art literature review and implement the selected approaches. (2) Based on the investigation results from (1), an important task is to analyze the strength and weaknesses of existing approaches and further optimize them to improve the recommendation performance. This includes evaluating the methods with real-world datasets.
Data Description: This project will use the Adressa Dataset, a Norwegian news dataset containing about 113 million events in connection with anonymized users over a 90-day period from 1 January to 31 January 2017. This dataset has been pre-processed and stored in the json-style format. The size of the dataset is around 16G. It is publicly available with the license CC BY-NC-SA 4.0.
Challenges (business and/or research): Different from other domains, e.g., music and e-commerce, recommending news articles to online users has been recognized as a challenging problem in both academia and industry because of several remarkable characteristics: short shelf lives, continuous, anonymous users with few user profiles available, large-scale and complex relations among news entities.
This is a joint project between NorwAI and Kavli Institute for Systems Neuroscience, NTNU.
The objective of this project is to develop a more objective and accurate method for estimating the sentiment load of each word in a sentence. This project will use signals driven from functional imaging to establish a mapping function between brain signals and sentiment, which can be used to estimate the perceived sentiment of each word in a sentence.
The project will involve the following steps:
1. Literature Review: A review of existing research on sentiment analysis and neuroimaging techniques, including MRI, will be conducted to identify the current state-of-the-art and the gaps in the field.a. Related reference (start point): Wankmüller, S. and Heumann, C., 2021. How to Estimate Continuous Sentiments From Texts Using Binary Training Data. In Proceedings of the 17th Conference on Natural Language Processing (KONVENS 2021) (pp. 182-192).b. Khosla, M., Jamison, K., Ngo, G. H., Kuceyeski, A., & Sabuncu, M. R. (2019). Machine learning in resting-state fMRI analysis. Magnetic resonance imaging, 64, 101-121.
2. Data Collection: A dataset of text inputs and MRI scans will be collected from a group of participants. The text inputs will be annotated with emotional labels to provide a training dataset for the machine learning algorithm.
Note: Now we are trying to find available dataset from the database. Otherwise, we will start to collect the dataset with the help from the Ziaei group, who are experienced in collecting this type of data.
3. System Design and Development: The project involves a functional imaging session during which they are presented with a large set of words, each carrying a positive, negative, or neutral sentiment. By analyzing the fMRI signals generated during the presentation of each word, we aim to cluster characteristics of the signals that correspond to each sentiment category.
4. System Evaluation: Once we have established a reliable mapping function between fMRI signals and sentiment, we will apply this function to estimate the sentiment load of each word in a sentence. This can be done by presenting the sentence to a participant in an fMRI session, one word at a time and analyzing the corresponding fMRI signals to estimate the perceived sentiment of each word.
Expected Outcome: The expected outcome of this project is a system that can estimate continuous sentiment based on textual annotated data and MRI with high accuracy and reliability. The system will be designed to be scalable and adaptable to various contexts, such as health care, social media, customer service, and marketing.
The potential supervisors are Lemei Zhang (lemei.zhang@ntnu.no) and Peng Liu (peng.liu@ntnu.no)
AI and Big Data technologies are central in many digitalization or digital transformation projects. They are sometimes used to cut costs of existing services, but the technology is often adopted to offer improved or brand new services. There are today best practices from many industry sectors, and practitioners are researchers have worked out guidelines for the integration of AI in organizations. However, many projects suffer from a lack of understanding how the nature of AI affects technical infrastructures and business strategies, and social and ethical implications are often underestimated and not sufficiently addressed.
In this project the students will investigate in more detail how AI creates value along the value chain. We will identify critical features of AI-based solutions and analyze how these features affect organizations’ business strategies, create opportunities or impose limitations. Central to the project is a solid understanding of the complexities of AI and what it takes for an organization to benefit from the technology. The intention is to work out detailed technical and organizational guidelines and checklists for leveraging AI-based solutions to create long-term business value. In-depth discussions with companies may be needed as part of this analysis.
NorwAI (Norwegian Research Center for AI Innovation) is a new research center on AI and Big Data. Its goal is to develop cutting-edge theories, methods and technology for efficient and responsible exploitation of data-driven AI in industrial innovations. The industrial partners of NorwAI are Schibsted, DNB, Telenor, Cognite, DNV, Kongsberg, NRK, Sparebanken1 SMN, Trønderenergi, Retriever, and Digital Norway.
The purpose of this document is to propose a research topic for master-level students interested in 3D deep learning that can have huge values for Cognite. The end goal of the project should be a publication as well as the solution.
In order to build a fully representative digital twin, it is required to transform the captured reality in 3D space, i.e. point clouds, into a semantic representation. Traditionally, this is done by first fitting the geometry primitives to point clouds and manually aggregating the fitted primitives to semantic meaningful objects. With recent advances in deep learning, one can directly learn the mapping between raw point clouds to segmented ones (semantic segmentation) or the mapping between point clouds and objects using bounding boxes (object detection). This is proven to be useful to detect cars, cyclists, and pedestrians in the autonomous driving setting.
However, those techniques are not directly applicable to large-scale industrial environments for the following reasons. For object detection techniques, such as PointRCNN, unlike in the driving scene where the aspect ratios of objects are within certain ranges, industrial objects can have an arbitrary length in one dimension (i.e. pipes) or have an arbitrary skeleton. Another problem to apply to the industrial setting is that the point clouds are on very large scales and the objects are densely connected. For semantic segmentation techniques, such as PointNet++, although it would be possible to segment the point clouds, the segmented point clouds still need to be further processed to reconstruct their geometry shapes through model fitting techniques such as RANSAC or its variants. However, in practice, this pipeline requires carefully choosing the right parameters for different datasets and different geometry primitives, thus hindering the possibility to scale.
In this proposal, we would like to explore supervised learning-based methods to directly detect semantically meaningful objects in large-scale process plants with geometry primitives beyond the bounding boxes. There are some attempts to fit geometry primitives to point clouds with supervised methods. However, their focus is on objects not on scenes and the primitives do not have semantic meaning. Nevertheless, their approach could be a good starting point combined with the standard bounding box-based object detection technique
Conversational agents (chatbots) are systems that can enter into dialogues with real users. Most current chatbots are rule-based, but some newer systems are based on generative models and machine learning.
We are working on a generative conversational system for DNB's customer requests. The system makes use of deep learning, with a dataset of chat logs from DNB. As part of this work, we are looking into how conversational agents may be incorporated into robots with fascilities for speech generation and speech recognition. An example of such a robot is the robot heads manufactured by Furhat Robotics in Sweden.
In this project the students will explore ways of incorporating conversational agents into robots. In particular, we need to investigate how the limited functionality of a conversational agent may be extended to serve as a full-fledged conversational partner with access to external knowledge and dialogue state information. The project requires a solid understanding of machine learning and interests in NLP.
Description in which company/unit the thesis will be placed: NTB (Norsk Telegrambyrå AS) is the national news agency in Norway, owned by the media industry. NTB is a member of international organizations of news agencies, and is considered one of the most innovative agencies in the business. A daily production of at least 150 news articles, 50.000 pictures added daily to our picture service and several services on sports data, picture storage, press release and other services, has been the basis of NTB during the last years. A groundbreaking work on picture recognition with AI has made us the leading actor in Scandinavia, with 15.000 persons now automatically recognized. Our work on automated article production has been ahead of the rest of the business in Norway. Our automated translation service between Bokmål and Nynorsk has changed the way we produce news and is a service used by media companies, governmental institutions and private businesses. We are using technology to change the way NTB can continue being the closest partner to our customers.
Problem Description: NTB is currently developing an automated text robot that generates daily articles about every open real estate transaction in Norway. Because of the sheer number of transactions, around 100 000 a year, we believe that utilizing natural language processing is a must to ensure that the articles are as rich, varied and readable as possible.
One problem could be how to add parts of or a whole sentence to change and enrich the language by using NLP. By this minimizing the need for hand coding the language variations.
A successful solution to this challenge should also be somewhat portable to other automated journalism projects while upholding NTB’s standards for precision and quality.
Thesis Description: Using AI, we wish to explore the possibilities of enriching automated articles with varied and colorful language following journalistic norms. By training the AI on journalistic articles written by NTB, we expect automatically generated articles that vary from one another in style, language and flow, while preserving good grammar journalistic guidelines.
At present these automatically generated articles are created using a template solution, with checks and variations to create relevant and varied articles for the data available. For the real estate robot, this means a lot of manual work as there are a large number of edge cases. Writing variations for each edge case increases the manual work drastically.
Using NLP we desire to be able to vary the language used without having to manually write multiple versions, letting an AI write parts of a sentence or potentially entire sentences.
Data Description: Available data will consist of templates for active template solutions for article writing robots. Datasets containing input data for the templates. These will where required be synthetically generated to anonymize sensitive data. And the candidates will have access to NTBs articles archive, which contains tens of thousands of articles, spell checked and written in the format of NTB style.
The main focus will be on the housing transaction robot and associated data. This dataset consists of housing and transaction information for every housing transaction that occurs in Norway. This data will be synthetically generated. The data content can vary in both completeness and in quality. Which leads to a multitude of variations and edge cases that need to be covered by the article writing robot.
The data will be synthetically generated data - mimicking actual real estate data from the Norwegian Kartverket, thereby ensuring compliance with guidelines for data protection and anonymization - as well as the templates we are currently using to generate articles, a vast collection of real news stories from NTBs archive. And if necessary, the articles automatically outputted by the templates. All as JSON files.
Size of the data depends entirely on your demands and wishes, as we will both draw from NTB’s massive article database as well as generate synthetic data on demand.
Challenges (business and/or research): The challenge of this project is not only to be able to vary language in robot generated articles but to also do so while preserving journalistic standards. This adds to language requirements but also to the balance requirements on produced articles. By using keywords you should be able to generate a text or parts of a text, that gives a logical meaning and understandable the content, and minimizes the need for hand coding to make the language fluent and varied.
Contact: Per Christian Evensen Helme (per.christian.helme@ntb.no )
As it is well known, recent deep learning methods allow unprecedented performance for image classification, semantic segmentation, and object detection tasks. When building object detectors for marine applications, it is particularly important to have training sets which do not only contain a sufficient number of relevant object classes (ships, piers, buoys, ...), but also show the different variations in lighting conditions, and weather conditions (fog, rain, snow, ...). As it is difficult, if practically possible at all, to record a sufficient number of such images or videos for training object detector networks, this project aims at artificially generating these variations of image conditions, in particular different illumination and strongly different weather conditions. Image augmentation is a well-known technique for synthetically expanding limited or imbalanced training datasets used in computer vision tasks to improve generalization performance and to avoid overfitting. Likewise, augmentations can be used for testing network generalization performance, by imposing varying augmentations, e.g. two-dimensional augmentations such as shifts, rotations, etc., as well as natural phenomena, e.g. snow, fog, lighting conditions, etc., or even adding synthetic objects to the scene. The former is straightforward using simple transformations in two dimensions, whereas the latter should imitate realistic, and highly stochastic, scenarios that in reality depends on the depth information of the scene. Nonetheless, this is often done by basic image manipulations imposed uniformly over the whole scene without considering depth information and may result in unrealistic augmentations not representing reality and could possibly produce erroneous testing performance. As an example, fog or rain may be observed only at a certain distance and/or area and affects only specific far-away objects, rather than all objects in the scene. The proposed research topic is therefore to investigate techniques for extracting depth information from images or video and taking advantage of this information to apply augmentations of natural phenomena in 3-Dimensions. For the extraction of this 3D information, one can use recent methods of single image depth estimation by deep neural networks, or "motion stereo" from a moving camera (in case of video sequences). Both approaches can also be combined. Therefore, the first partial goal of this project is to explore methods for depth estimation for given training images or videos. The second goal is to use the extracted estimate of the depth structure to simulate the desired weather and illumination situation, such as fog, rain, or snow. Again, this can be done by "classical" physics-based computer graphics modeling, and/or by suitably trained deeep neural networks. Another important aspect is to benchmark 2D and 3D synthetic augmentations against images containing actual natural phenomena’s, i.e. network performance on synthetic scenes against true scenes. This project will be performed with the Norwegian company DNV GL which is interested in quality assurance of marine detection systems, and will also be related to the marine multi-sensro simulator project currently pursued at NTNU. Students who are interested in this project are advised to address Prof. Rudolf Mester at IDI for more details. The co-supervisors will come from DNV and probably also from the Cybernetics Department (ITK).
This project builds on a successful Masters student project in the spring term 2022 which achieved already substantial results related to fog and snow simulation. The new project will refine these approaches, match the synthetic weather effects to an available large database of scenes taken from long term recordings of outdoor webcams, and is expected to approach a still higher level of realism, possibly also targeting also synthetic video generation.
More than 80% of our global goods are transported by ships. Like the goods they transport, ships will eventually become waste and need to be broken down properly. The process of breaking down a ship involves a lot of people working at different parts of the ship. Currently, ship-breaking workers rely on noise to detect if there are other workers nearby. This practice is unsafe, since there is a possibility that the noise is unheard, and workers may not realize there are others nearby when performing the cutting process. Therefore, there is a need for visualizations that show the ship-breaking status and the presence of ship-breaking workers on-site. Such visualizations would be useful to improve safety in ship-breaking activities, as workers could check before cutting anything.
Key activities in this project will include:
This project will be carried out as part of the SHEREC project, which aims to improve safety in the ship-breaking process through digitalization and deployment of robots. The students will receive support from the partners affiliated with the project. In addition, the necessary 3D models will also be provided.
The project will be co-supervised by Dr. Taufik Akbar Sitompul (Department of Design, NTNU).
The main goal of this Thesis is a 3D-to-2D Face Recognition approach based on a combination of classical deep learning networks and the new trend in deep learning branch, Geometric Deep Learning Networks.
By the term 3D-to-2D Face Recognition (3D-2D FR) we refer to the procedure of face recognition using as input a typical 2D facial image query (or a sequence of 2D facial images of the same identity) against a dataset containing 3D facial models (i.e., facial meshes or point clouds). To the best of our knowledge, most of the 3D-2D FR state-of-the-art methods do not use deep learning procedures [1,2]. Thus, as a proposed approach, classical deep learning networks [3], able to extract meaningful features from sequences of 2D face images, as well as modern geometric deep networks [4], able to extract meaningful features from 3D facial models, could be combined, using appropriate loss functions [6], in the form of the so-called Siamese Networks [5].The training of such a network could be performed by using synthetic data to be generated by the FaceGen [7] tool that is available at the Visual Computing Lab.
======================References---------------1. https://www.sciencedirect.com/science/article/pii/S1077314216300480 2. https://ietresearch.onlinelibrary.wiley.com/doi/10.1049/iet-bmt.2015.01203. https://www.deeplearningbook.org/4. http://geometricdeeplearning.com/5. https://www.cs.cmu.edu/~rsalakhu/papers/oneshot1.pdf 6. https://www.sciencedirect.com/science/article/pii/S00313203193006037. https://facegen.com/
===================================== Requirements:
[Necessary]
Knowledge: Python (for the machine learning part). C/C++
Courses: TDT4195 (Visual Computing Fundamentals), TDT4230 (Graphics & Visualization), or equivalent.
A strong interest in Visual Computing and Machine Learning.
[Desirable]
Knowledge: Matlab, Machine learning environments, OpenGL.
Courses: TDT4265 (Vision), or equivalent
Supervisors:
Dr. Antonios Danelakis, Postdoctoral fellow, IDI, NTNU antonios.danelakis@ntnu.no
Prof. Theoharis Theoharis, IDI, NTNU theotheo@ntnu.no
Main goal:
Use a game engine for simulation with ROS2. The integration of a game engine and ROS2 allows us to leverage the power of the game engine to create realistic USV simulations.
The project work can be extended into a master thesis by implementing and investigating USV validation and verification methods within the developed simulation framework. A possible extension may also involve the utilization of the same simulator framework for data generation, data augmentation, and testing of situational awareness and machine learning methods.
Project description and tasks
● Review game engines including Unreal Engine 5 and Unity, and select the most favorable for integration with ROS2 and Otter USV's existing onboard and simulation systems
● Integrate the chosen game engine with ROS2. Create a ROS2 node and integrate it with a game engine project by setting up communication between the two platforms. It is possible to use the default communication system in ROS2 (DDS - Data Distribution Service) through a fast DDS implementation or by directly utilizing the shared memory protocol in fast DDS.
● Set up Otter with the SeaSight Camera and Lidar systems, and set up a simulation environment using the game engine (can be based on an existing setup in Gazebo)
● Transfer camera images from the game engine to ROS2 as a ros topic. Transfer Lidar pointclouds created based on objects setup in the game engine to ROS2 as a ros topic
● Analyze the setup and simulation results in relation to real-world data and existing results in the literature
● Write a report
about Maritime robotics
Maritime Robotics is a leading provider of advanced autonomous technology, enabling safe, sustainable and cost effective maritime operations and applications. Since 2005, Maritime Robotics has developed and delivered Autonomous Navigation Systems and Uncrewed Surface Vessels (USVs) to customers worldwide, enhancing the operational capabilities for various maritime applications.
The following project and thesis proposals aim at research and development of advanced control, situational awareness, and simulation systems that enable remote and autonomous operation of surface vessels.
In 2019, the healthcare region of Central Norway procured an Electronic Health Record (EHR) system from an American vendor (Epic Systems) to fulfill the national vision of “one citizen, one journal” – the project is called Helseplattformen (HP). In 2022, HP implemented an integrated EHR system connecting primary healthcare providers in Trondheim municipality and secondary providers at St. Olav’s Hospital.
In this project, we plan to do a follow-up study of the use of the HP system by healthcare professionals in their practices by building on the earlier results. This study uses qualitative research methods to address questions like why we have too much data in the HP system, who provides it and why, who uses it, and for what purposes and what values are achieved. The candidate will do initial literature studies on the topic and design a case study with data-collection methods like observations, interviews, and archival data. The collected data will be analyzed qualitatively to define a coherent concept explaining the role of data and its value in healthcare professionals’ use of the HP system. The subject areas for this project include information systems, computer-supported cooperative work, and human-computer interaction.
This task requires that you have a good understanding of, and are interested, in empirical qualitative research. Please contact Babak before you select this task.
Working language for this task is English. The thesis can be written in Norwegian or English but we recommend English.
See this link to learn more about our group and an overview of other tasks from our group: https://digipub.idi.ntnu.no/en/available-msc/
In this project, we plan to study the use of Microsoft Copilot by computing students at the NTNU using qualitative research methods to gain a rich understanding of the phenomenon. The candidate will do initial literature studies on the topic and design a case study with data-collection methods like observations, interviews, and archival data. The collected data will be analyzed qualitatively to define a coherent concept explaining the practices of computing students using Copilot. The subject areas for this project include information systems, computer-supported cooperative work, and human-computer interaction.
Introduction
Amidst a burgeoning aging populace and surging demand for homecare (hjemmetjenester) services, there emerges an urgent call for innovation and optimization in service delivery. Decentralized technology presents distinct advantages, such as user-centric data ownership, accessibility, heightened privacy, security, and transparency. These attributes hold the potential to significantly enhance the efficiency and efficacy of municipal homecare services. The proposal aims to explore how decentralized technology could transform and enhance homecare services administered by Trondheim municipality, heralding a paradigmatic shift in service provision.
Central to the project's focus is the introduction of a decentralized, user-owned health wallet platform to tackle the myriad challenges facing contemporary homecare services. This pioneering platform empowers individuals to assume control over their health data, enabling secure storage, management, and sharing of medical information according to their preferences. Against the backdrop of mounting concerns surrounding privacy breaches and data mishandling, this initiative offers a compelling alternative by reinstating ownership and oversight of sensitive health data to patients, thereby fortifying security and privacy protocols. Moreover, it promises to foster transparency and bolster patient autonomy, fostering active participation in their healthcare journey. Note that this isn't just about data – it's also about empowerment of patients!
Furthermore, the envisioned platform holds the potential to revolutionize not only homecare services but also broader healthcare provision by facilitating seamless data exchange among patients, healthcare providers, and other stakeholders. Such enhanced connectivity ultimately promises improved outcomes and a more patient-centric healthcare ethos.
Tentative tasks
Training of AI agents in virtual environments using deep reinforcement learning.
Some examples of virtual environments that we are using: CARLA, NVIDIA Omniverse (e.g. Isaac Sim or DRIVE Sim) and MARS, etc.
Some references to SotA approches that we are looking into:
Additional industri references:
We are also working to create a realistic virtual replica / twin of the area around the campus (Gløs) to use in simulators like CARLA and DRIVE Sim for training of AI agents in a local environment.
Task: Develop an app for mobile devices, that can be mounted in public transportation like busses, can access the camera of the device as well as relatively cheap and accurate positioning equipment with CPos corrections (cm accuracy) and have AI models for assessing and geo-referencing the condition of all road objects visible from the road (one application, other applications could be to create and update HD-maps, match real-time images to a reference for back-up localisation, collect data for neural rendering etc.).
Collaboration: the project will be a collaboration between the national road and mapping authorities, SINTEF and several counties / municipalities. Huge innovation potensial.
Various aspects related to AVs, including the use of NAP-lab's new research platform for AD.
End-to-end approaches to AD like imitation and reinforcement learning.
Modular approaches to AVs (i.e. mapping and localization, perception and prediction, planning and control), including the use of the functonality provided by the NVIDIA DRIVE software stack and DRIVE / ROS integration.
The use of simulated environments (like CARLA and NVIDIA DRIVE Sim), new ways to generate a realistic virtual environment using real world data (e.g. NeRF), and domain adaptation from simulated to real world scenarios.
AVs as mobile sensor platforms (assessing the condition of road objects to update Digital Road Twins) and AD in a challenging nordic winter environment (both in collaboration with SVV)
Point Cloud processing (real-time detection and tracking of objects like pedestrians and vehicles in LiDAR data, generation of point-cloud references (e.g. SLAM), and updating this reference as well as re-localize relative to it, collaboration with karverket).
Remote control of an AV using 5G (e.g. management of an AV fleet)
+++
If this is a topic of high interest to you the most effective way would be to have a talk and see if we can find a project highly motivating for you at the same time as the overall NAP-lab project moves forward.
Main investigators for the NAP-lab project is Frank Lindseth and Gabriel Kiss
Additional info
Progresso is a programming tutoring system that provides learners with personalised courses from various domains. Currently, it offers a Java programming course with interactive third-party material. The system provides infrastructure for collecting and displaying different learning analytics, logging learners' activities, customisation, and authentication of users.
The aim of the thesis is to continue development on the system, provide new personalisation and customisation options, add access to third-party content providers (Java, Python, and SQL courses), create a framework for adding new courses, and allow standardised integration in different LMSs. The students are expected to execute a research study to evaluate implemented personalisation features of the system.
In an era marked by increasing digital transactions and online interactions, ensuring the security and integrity of personal identities has become paramount. Traditional methods of identity verification, such as passwords and biometrics, are often susceptible to fraud and exploitation. However, the integration of Artificial Intelligence (AI) offers a promising avenue for strengthening identity security measures. By harnessing AI algorithms for identity verification, organizations can enhance accuracy, efficiency, and resilience against fraudulent activities. This proposal seeks to explore the implementation of AI-driven identity security systems to fortify the protection of individuals' personal information and prevent identity theft.
Tentative Tasks:
Affine transformations are in the heart of Visual Computing and a very common topic in examinations. This project will look at ways of generating random affine transformation questions for examination settings. To be more generally applicable over the internet, it is intended to implement it in WebGL.
Requirements:
Knowledge: OpenGL
Supervisor:
The vehicle industry, as well as software and hardware providers are rapidly developing sensor systems and artificial intelligence (AI) methods for sensing the road environment. Connected and Automated Vehicles (CAVs) are argued to have a large potential for accelerating traffic safety and efficiency. Digital twins allow not only to visualize how things work, but also simulate various future scenarios. This is particularly interesting for autonomous vehicles which can be trained in a simulated environment. Furthermore, changes to the algorithm can be validated in a digital twin before deployed on the vehicle. Building a digital twin of a nordic environment allows for development of AI techniques designed for such an environment.
Possible topics:- NeRF and Gaussian splats: create local environments based on data acquired with an autonomous platform. Dynamic environments that take into account vehicles, pedestrians and cyclists (e.g. MARS, StreetGaussians)
- Underwater NeRFs for representing shipwrecks and other underwater artefacts.
- Digital twins visualization: extend the currently available DigitalTwin of Gløshaugen area and make it more realistic. Final goal is to import it into Nvidia Omniverse so it is usable to train a network that is designed for our autonomous vehicle.
- Nvidia CloudXR: visualize a digital twin in VR/AR and simulate various driving conditions
Supervisors: Frank Lindseth, Gabriel Kiss (COMP/IDI)
The focus of this thesis is to develop an Artificial Intelligence based system to help the students learn mathematical concepts while playing educational games. One of the ways to provide help is to find out the difficult moments during the interaction and then supporting the students when they are faced with such moments. The challenging aspect of such projects is the “cold start problem”. We need to know in advance how to detect the difficult moments for individual students. Solving this problem will be a key aspect of this thesis
Thesis DescriptionIn a first step, the student(s) will design and implement the feedback tool using the wearable sensors. Afterwards, they will conduct a user study in order to test the usability of the system with a number of students. Once the usability of the system is established (with the last changes in the system), the student(s) will conduct a larger user study to evaluate the effectiveness of the system. Finally, the candidate(s) will analyse the collected data and write up his/her thesis.
RequirementsThe ideal candidate will have a background in system design and basic machine learning. Solid programming skills and an interest in hands-on development and experimentation is also a requirement.Programming skills: Python/Java.
Supervisors: Ilias Pappas
The recent advances in machine learning (ML) and artificial intelligence (AI) offer an opportunity to design and build better tools as solutions for existing societal challenges. Such solutions can contribute towards dealing with the United Nations’ 17 Sustainable Development Goals (SDGs).
The project consists of a literature review in the area which will help narrow down the topic. This will directly impact the empirical part of the thesis, that will also be based on the students’ interests. The second phase includes designing an empirical study and collect data employing qualitative and/or quantitative methods. For example, the project may focus on AI tools that deal with prediction of crisis events or disaster management. In the final phase, the students will analyse the collected data and write up their thesis.
Relevant literature:Tomašev, N., Cornebise, J., Hutter, F., Mohamed, S., Picciariello, A., Connelly, B., ... & Clopath, C. (2020). AI for social good: unlocking the opportunity for positive impact. Nature Communications, 11(1), 1-6.https://www.nature.com/articles/s41467-020-15871-z
AI has leapfrogged many areas, including computer vision, enabling open discovery scenarios in Augmented Reality (AR) in Education. The thesis will replicate existing examples using openAI’s CLIP service (or YOLOv5) and evaluate its applicability for open discovery processes in an AR learning application. CLIP (or YOLOv5) are used to detect real-world objects in photos of the user’s physical environment, to enable scenarios like content search.
With this master thesis project, you will:
* Design and develop an architecture for processing, detection, and similarity search of real-world objects from submitted photos
* Assess its feasibility and evaluate efficacy using prepared small test collections in a so-called ‘pseudo experiment’
* Investigate potential usage scenarios through proof of concept implementation (e.g., visual search, environment tagging, interaction triggers), which can be tested with users in educational scenarios (e..g, vocabulary learning or industrial workplace training)
Outline solution:
* submit viewfinder image to discovery service to identify objects
* start object search using AI computer vision services (see the CLIP Jupyter notebook example below)
* return list of entities identified and their boundary box info in the image
* store data about recognized entities, together with session information
* retrieve sessions with same or similar entities
* test adaptive AR tracking with clipped image as image target
Aim is to interface this open discovery service with MirageXR (https://github.com/WEKIT-ECS/MIRAGE-XR/), the AR learning experience editor and player, to support discovery of learning content that is available to the users in their direct, physical surroundings. This can be used to, for example, submit a photo of a room containing an Internet router, an ultrasound machine, a picture of a cat, a coffee mug, a flour package, returning experiences that have been tagged with these objects.
Type vs token:
* detected objects can be generic (e.g., "dog", “book”, “milk carton”)
* Or specific (e.g., specific router of brand + model number)
How to:
* Here is a basic pipeline with openAI's CLIP https://www.pinecone.io/learn/series/image-search/zero-shot-object-detection-clip/
* And there are yolov5 based alternatives (but they require hunting down a training database or use a pretrained free model, same as LLMs)
The students will have access to a very well-equipped IMTEL VR lab (https://www.ntnu.edu/imtel/) containing various modern AR and VR devices and laptops.
The target devices for the project are Apple Vision Pro, HoloLens 2, and Oculus Quests 3.
For any questions about the task, please, contact Mikhail Fominykh mikhail.fominykh@ntnu.no.
Prof Dr Monica Divitini, Professor at the Department of Computer Science, NTNU
Dr Mikhail Fominykh, Researcher at the Department of Education and Lifelong Learning, NTNU
Prof Dr Fridolin Wild, Professor AR/VR at the Open University, United Kingdom
This project/master thesis will build on existing bulk of knowledge about Artificial Intelligence (AI) and diversity in software development to provide increased knowledge and solutions about how learning AI can empower groups that traditionally are minorities in AI, such as women, immigrants, people with disabilities. Specifically, in this project/master thesis, the student(s) will:
- design a knowledge package including digital an in presence innovative course
- contribute to developing an inspiration academy to facilitate inclusive education and empowerment with AI
- design, implement, and evaluate new tech solutions which contribute to empower diverse groups with AI
The supervisor will provide the student(s) with initial literature and help the student(s) to access to stakeholders and initial data for the research Resources
sbs.idi.ntnu.no
women-stem-up.eu
eugain.eu
https://www.europarl.europa.eu/topics/en/article/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence
This project works with diagnostics of eyesight issues and spectacles. It is carried out in close copperation with an international start up working with new ways of addressing eyesight problems.
If you are interessted in computer vision, eyesight, industrial applications and impact in the 3rd world, this is a project for you.
Please contact the supervisor before applying.
It is well known that producing software for knowledge work is challenging. Knowledge can be tacit, social, produced through negotiation, and emerges in practice.
However, with the emerging AI solutions, such as generative AI, we see more widespread use of AI in knowledge work, across industries, such as fisheries, police, and the justice system.
A critical component for successful implementation of AI is understanding how AI works in practice, and how AI supports particular forms of knowledge work. Certainly, there are many forms of knowledge work, and as with any tool, tools must be aligned with and work practice.
The project will start as a literature review, but can be extended with an empirical study for a master thesis project. A potential case is DNV (dnv.com) and their initiatives to support their experts with AI tools.
The project will be done together with SINTEF Digital and a case company
Besides the more or less fixed topics which are found in the list, there is always the possibility to define a project according to your own interests, as long as it is scientifically solid or a real engineering challenge.
Topic areas:
* Certifable AI for safety critical systems
* Efficient deep learning architectures
* Structured Deep Learning: implanting physical and semantic knowledge in deep learning architectures
* Computer Vision in bad visibility situations (road traffic, underwater robots, ...)
* Mobile Systems (cars, ships, drones, indoor robots): building up prototypes
* Autonomous Racing Cars (e.g. in the Revolve context)
* Machine Learning for Planning and Control in Mobile Systems
* Merging Statistical Pattern Recognition and Deep Learning
* Machine Learning and Computer Vision for Future Smart Traffic Infrastructure
For those who are interested in industry-relevant topics: We are cooperating with leading industrial companies, both in Norway as well as in Germany and Sweden, for instance on Qualification of AI-based Systems, Drone-based Inspection, and have ongoing cooperations with leading firms in the Autonomous Driving area.
The green shift is high on every executive’s agenda, and with good reason. The urgency of the climate crisis and associated transition to a sustainable society changes the way firms create, capture, and deliver value. Shifting the very fabric of today's business landscape. Firms must now deliver on a triple bottom line (environmental, social, and economic) and not only meet today's needs from customers and shareholders, but also future generations' needs and opportunities for value creation. A strategic response is required, and firms must make structural changes to accommodate a fully sustainable business model (SBM). Research suggests that firms that manage and mitigate their exposure to climate-change risks while seeking new opportunities for sustainable value creation will generate a competitive advantage over rivals in a carbon-constrained future. However, transitioning towards a SBM is challenging and companies often lack the necessary data and insight to make correct and effective business decision. Artificial Intelligence (AI) offers a possible solution by establishing a basis for data-driven and fact-based decision making. This makes it easier for firms to take a systems perspective, quantify impacts, and reduce the complexity of the sustainable transition. Although real and theorized examples of AI enabling SBMs exist, a comprehensive understanding of the relationship between AI and SBM is still missing, leaving a gap in our understanding of the underlying mechanisms and inhibiting firms’ ability to accelerate their sustainable transition. Thus, this project aims to take stock of current knowledge by studying the following research questions:
RQ1: What do we know about the relationship between AI and SBM? RQ1.1: How can companies leverage AI for SBM? RQ1.2: How can the relationship between AI, SBM and competitive performance be conceptualized?
Even though oceans are very important for human life and societies, we have very little understanding of marine ecosystems which are very complex systems. Ocean observatories and other underwater monitoring systems provide data streams that cover physical, chemical and biological ocean properties.
Marine animals are very vocal; many invertebrates, fishes, and nearly all marine mammal species produce sounds. Underwater acoustic data provide scientific insights in a broad range of fields including animal vocalizations (biophony) and anthropogenic noise (anthrophony) in the marine environment. However, underwater sounds are very challenging to identify and classify due to the variability of sound events and conditions under which they were recorded. This creates interesting research challenges to deal with this variation, sparsity and noise.
This project aims to develop effective classification and analysis models of large acoustic data streams and additional data from ocean observatories.
The project will run in collaboration with the Department of Biology at NTNU.
This project utilizes data science techniques to visualize and analyze open-text responses from students. We actively aim to gain valuable insights into students' perceptions in educational contexts by employing sentiment analysis. This will enhance our understanding of student experiences and actively inform strategies for improving teaching and learning practices by exploring innovative methodologies. The project endeavors to transform raw open-text responses into actionable insights by actively utilizing data science techniques such as natural language processing and machine learning algorithms. The dataset collected from student responses during lectures is expected to be substantial, potentially ranging from a few to hundreds of responses per question. This active engagement with cutting-edge techniques empowers us to extract meaningful insights that can inform tangible strategies for enhancing teaching and learning effectiveness.
Imperfect Information Games are those in which players only know some but not all aspects of the state of the game. Texas Hold'em poker is a classic example: each player knows their own two hole cards and the public cards (face-up on the table), but does not know the opponent's two hole cards. Today's AI poker systems typically employ a combination of tree search and neural networks, where the latter serve as mappings, for example, from probability distributions over hidden states to winning probabilities. These networks greatly reduce the size of the search tree.
In this project, the student(s) will choose an imperfect-information game of their choice. The game should be relatively simple to simulate, hence preferably a board or card game, and apply some of the same techniques found in the AI poker literature, such as this seminal paper on DeepStack:
"DeepStack: Expert-Level AI in heads-up no-limit poker", Science, 356, pp. 508–513 (2017).
Ambitious students can choose Texas Hold'em or any other version of poker for this project as well. Various economics scenarios, such as auctions, may also be relevant.
This project requires a strong interest in AI programming and in game theory.
This preproject has been listed here on the background of topic discussions with two students. We make a joint decision on the particular topic direction, either aiming at vision-based salmon identification and tracking, or at a perception system for autonomous ships. In both cases, the preproject will be focused on exploring the literature and the technical background for both target MSc projects, and the main goal of the preproject is, besides acquiring the necessary background knowledge, to set up a detailed research plan for the subsequent MSc project on the jointly agreed topic.
This project is offered in the context of a big NFR-funded research project which is performed as a cooperation between the cybernetics department (ITK) and computer science (IDI). The overall project AROS deals with the task to provide autonomy to the snake underwater robots developed by ITK (Prof. Kristin Y. Pettersen and colleagues).
In this context, the task to be addressed in a student project (pre-project or masters thesis) is to explore different approaches for underwater computer vision under the particular conditions that a underwater robot will encounter:
* reduced range of visibility due to limited available illumination and limited transparency of the water
* visual clutter due to fish and plankton
The overall task in this setting is to generate a dynamical (situation dependent) model of the environment around the snake robot, and to estimate its ego-motion.
The project will employ / explore both classical computer vision methods as well as modern methods from machine learning / deep learning. The goal is to empirically evaluate the suitability of several candidate methods under the challenging visibility conditions given for this underwater scenario.
Co-supervisor: Ahmed Abouzeid (postdoc in AI group, at NTNU)
Artificial Intelligence (AI) has increasingly been used for making decisions with high impact on people’s lives. Some examples include decisions about hiring, medical diagnosis, and surveillance systems. The ethical dilemmas these processes give rise to have yet to be solved. Issues around bias and unfairness, transparency of the process, and explainability are the main challenges we face. Similar to GDPR that focuses on privacy issues, other regulations covering the rest of ethical AI problems are currently being intensely worked on [1]
The objective of this master’s thesis is to investigate how to ensure fair decision-making in AI-based hiring systems, more specifically in the first scanning and shortlisting of applicants. This work is connected to our work on BIAS—a research project on revealing and mitigating bias in AI-based hiring systems (EU link of the project: https://www.biasproject.eu/).
Currently, mainstream Machine Learning (ML) methods have been the mainline approach used in decision-making. Due in large part to the black-box nature of ML methods [2], and quality issues around training datasets, this approach has led to serious violations of ethical and legal standards [3].
BIAS has adopted a “similar individuals should be treated similarly” approach to fairness, where objectivity and consistency are core principles. This view of fairness is not new—it can be traced back to ideas expressed by Aristotle. BIAS is also investigating transparency in AI-based decisions—when is it possible to break down the decision-making process into smaller components that can be managed and evaluated separately.
At the core of our notion of fairness is how we define similarity. In a hiring context, similarity between candidates features such as age, gender, education level, and work experience.
When a human is behind the process, they generally weight these features differently, and these weights are dynamic—they change across hiring scenarios. The same HR professional may weight work experience as very important for a senior level role, for example, while weighting gender more for a factory worker position. For an AI-based hiring system, a key challenge lies in deciding the relative degree of importance of features to ensure objectivity and consistency with past decisions in a hiring company. BIAS is working closely with HR professionals to elicit weight values from them directly. We are also automatically extracting this information from the past/training data. This master’s thesis focuses on the automatically extracted weights of features importance, and will compare these results with the human provided weights.
Data:
Recruitment dataset (labeled) created for the BIAS project is made GDPR compliant and may possibly be used in the work for this thesis.
Required Skills:
Useful to have (optional):
References:
[1] Narayanan, D., Nagpal, M., McGuire, J., Schweitzer, S., & De Cremer, D. (2024). Fairness perceptions of artificial intelligence: A review and path forward. International Journal of Human–Computer Interaction, 40(1), 4-23.
[2] Rueda, J., Rodríguez, J. D., Jounou, I. P., Hortal-Carmona, J., Ausín, T., & Rodríguez-Arias, D. (2022). “Just” accuracy? Procedural fairness demands explainability in AI-based medical resource allocations. AI & society, 1-12.
[3] Muckley, E. S., Saal, J. E., Meredig, B., Roper, C. S., & Martin, J. H. (2023). Interpretable models for extrapolation in scientific machine learning. Digital Discovery, 2(5), 1425-1435.
The focus of the thesis is to improve and test an existing intelligent feedback system that helps the students while they are programming. This help should be provided in real-time using the eye-tracking data from the student and the log data from the IDE that the student is using. The challenge is to develop a system that is both effective and efficient in helping the students when they are facing difficulties in programming medium-size software.
Thesis DescriptionIn a first step, the student(s) will design and implement the gaze-aware feedback tool. Afterwards, they will conduct a small user study in order to test the usability of the system with a small number of students. Once the usability of the system is established (with the last changes in the system), the student(s) will conduct a larger user study to evaluate the effectiveness of the system. Finally, the candidate(s) will analyse the collected data and write up his/her thesis.
RequirementsThe ideal candidate will have a background in basic machine learning and system design. Solid programming skills and an interest in hands-on development and experimentation is also a requirement.Programming skills: Python.
This project involves the utilization of advanced tracking and computer vision techniques like graph neural network, convex optimization, and self-learning to monitor and analyze the behavior and interactions of multiple animals in indoor settings, with the specific aim of supporting selective breeding efforts. Selective breeding is a process in which animals are chosen for reproduction based on desirable traits to enhance specific qualities within a population. In this context, the project seeks to enhance the selective breeding process by applying technology-driven approaches to gather data and insights about the animals' behavior and characteristics.
The intersection of artificial intelligence (AI) and healthcare presents an opportunity to enhance medical diagnostics, improve patient outcomes, and streamline the work of healthcare professionals. This master project proposal invites students to contribute to this transformative field by developing an innovative web application. This application will leverage AI technologies to analyze medical data, offering insights ranging from data visualizations to complex diagnostics. The project integrates three key areas: scintigraphy image analysis, blood analysis data, and anamnesis analysis. However, students can choose the direction of their studies and focus on the area(s) of their interest.
RBK (fotball analytics):
Video presentation
Presentation with sound-clip embedded in each slide (download and listen)
St. Olav / NTNU Med. fak. (medical image analysis):
Abdominal Aortic Aneurysms (AAA): Is Minimal invasive or open surgery best for a given patient. (video)
Abdominal Aortic Aneurysms (AAA): Is Minimal invasive or open surgery best for a given patient. (text)
Brain segmentation from high-res MR images
Kartverket:
Video (for the three projects below)
Bruk av kunstig intelligens til klassifisering av laser punktsky fra flybåren datafangst
Beregning av DOP-verdier for å predikere GNSS målekvalitet
Strømming av punktsky fra database til web viewer
NINA:
Monitoring Norwegian nature loss with satellite-based earth observation and AI (Video)
Monitoring Norwegian nature loss with satellite-based earth observation and AI (Text)
MIA Health:
MIA (Monitorering-Innsikt-Aksjoner): Dine data. Din digitale tvilling. Din helse (video)
MIA (Monitorering-Innsikt-Aksjoner): Dine data. Din digitale tvilling. Din helse (tekst)
Catchwise:
Catchwise: Predicting where to find fish for commercial fishing boats
Maritime Robotics:
Dense Monocular Depth Estimation for Unmanned Surface Vessels
The emergence of whole slide imaging technology (WSI) allows for digital pathology diagnosis. The applications of digital pathology are expanding, from lesion detection and segmentation, to quality assurance and prognostication. The specific application in this project is related to lung cancer staging and is a collaboration with St Olavs Hospital and Levanger Hospital. A relevant topic will be to develop and validate ML techniques for automatic assessment of WSIs from established, well-described cohorts of lung cancer patients.
Tasks:
Ultrasound is becoming the imaging modality of choice for cardiac interventions. During cardiac surgery the location of instruments, as well as anatomic landmarks is crucial information for the surgeons. Today most of these tools are localized manually or semi-automatically, however automating them would improve the accuracy and patient safety.
Possible topics:
- image denoising via diffusion models
- Generative models or GAN's for multi modal image synthesis
- AI for aortic valve detection in mitral trans esophageal acquisitions of the heart.
-3D segmentation of valves, full 3D segmentation of the mitral valve from echocardiographic data. U-Net or more advanced architecture to be considered
-3D tracking of structures of interest in echo recordings, visual transformer-based architectures to be investigated, temporal consistency to be enforced
An immune system approach to fake news classification is currently under development . It is an exciting new approach to Fake News classification, drawing inspiration from antibody and antigen concepts from nature. This project seeks to extend and refine the current approach in various ways. The student(s) may choose their own path.
In a digital age, where users are faced with a significant amount of data, the recommendation system has become an essential part of everyday life to assist users in selecting products and services that are suitable for their needs. The music industry, in particular, has seen a rapid growth in the recent years, with users now able to access a greater variety of content. The need for user-tailored recommendations is, therefore, a necessity. An ongoing project applies the AIRS immune system algorithm to content based filtering for Automatic Playlist Continuation (APC). This project can further explore the application of AIRS or a different immunt system approach for music recommendation and can extend content-based to content-based and collabroation-based.
Description in which company/unit the thesis will be placed: Aneo is a renewable energy company with near 300 employees and headquarters in Trondheim. Aneo aims to produce more renewable energy and develop new, unique products and services to promote further electrification and efficient utilization of renewable energy. AI plays a key role in achieving this goal due to the increasing need for automation and data- driven decision making. We have extensive experience supervising students with weekly meetings and access to AI and domain experts.
Problem Description: Aneo owns and operates 11 wind farms with 225 wind turbines. An important task inoperating wind parks is to forecast how much energy will be produced hourly the next day, socalled day-ahead forecast. This forecast is used to balance electricity production with consumption to maintain stable power grid frequency. The accuracy of the forecast directly influences the profitability of the electricity producer.
Thesis Description: Aneo uses ML to forecast wind power, with weather forecasts as the main input. Weatherforecasts contain multiple weather scenarios, so-called ensemble. Using multiple scenarios as inputs rather than one (the most probable) might improve theaccuracy of the wind power forecast and estimate uncertainty. We want students to find a good way to utilize weather ensembles for probabalistic windpower forecasting.
Project tasks include: - Data preparation and exploration - Building models to forecast wind power production - Conducting experiments to evaluate and compare models - Analyzing experiment results
Description in which company/unit the thesis will be placed: Aneo is a renewable energy company with near 300 employees and headquarters inTrondheim. Aneo aims to produce more renewable energy and develop new, unique products and services to promote further electrification and efficient utilization of renewable energy. AI plays a key role in achieving this goal due to the increasing need for automation and data- driven decision making. We have extensive experience supervising students with weekly meetings and access to AI and domain experts.
Problem Description: Aneo Retail optimizes energy consumption and service costs for cooling systems in retailstores. Part of our work is to detect problems with the equipment in the retail stores and schedulereparation and maintenance. To make this process more efficient we want to to predict faults in advance. This would allow us to optimize maintenance scheduling, order parts and schedule work in away that minimizes consequences.
Project tasks:
• Data preparation and analysis. • Building models to predict future faults, time until faults. • Conducting experiments to evaluate models. • Analyzing experiment results.
Data Description: We will provide two types of data:
We have hundreds of fault events and millions of measurements. All data will be provided in CSV format. The data should be kept in private. We provide cloud infrastructure necessary for the project.
Description in which company/unit the thesis will be placed: Aneo is a power producer with both hydro and wind plants – now we will do more even on wind. Aneo is already a major player in renewable energy for a better society. As the owner and operator of the power plants, Aneo takes care of renewable energy to monitor, control,and optimize the performance of the generation or transmission system or added values through system of computer-aided tools. Among which nowadays AI plays an even more important roles in such system in order to make better decision under increased complexity and risk. AI team in Aneo is responsible for providing the innovated, reliable and robust AI energy system for both Aneo and the third parties, and coordinates Aneo’s AI activities and membership in Norwegian Open AI Lab (NAIL) and participates in research work of the recently established Centre for Research-based Innovation NorwAI.
Problem Description: Operating wind turbines in remote locations is challenging. Aneo experienced bushings,bearings and gearboxes in some wind turbines fail earlier than expected. Such failures can be detected through trend analysis of vibration data and/or other types of data, but the identification of the root cause is a very manual process. The objective of the project is to automate the identification of the root cause to support maintenance decision making for operators. The faults happen frequently enough that the farm owner will benefit significantly from an automated analysis process, but there may not be enough cases for supervised learning to be relevant currently. However, all existing failures have expert description which can help the students better understand the problems. The students are encouraged to use any available technologies to automate the process.
Thesis Description: The project aiming to solving the following tasks:
Data Description: The project is based on existing vibration data from wind turbines. These consist of short periods of high frequency data as well as statistics. Some faults have already been identified in the data. If needed, SCADA data from wind turbines are also available. The data needs to be kept internally.
In the era of Open Date, many parliaments publish their proceedings digitally. The transcribed speeches represent a valuable resource for political education. Social Sciences have invested efforts to analyze these text manually. AI promises to auomate analysis. Most machine learning methods require labeled data to find relevant patterns. These annotations are lacking at the moment.
The thessi will develop a way to efficiently annotate Norwegian political texts. A thesis in the Spring term 2022 has already laid the groundwork in preparing a graphical user interface for annotations. The main research questions concern the efficiency of annotating and motivating annotators. The thesis will explore ways to ease the annotation process with the help of suggestions. The candidate(s) will train a recommender system to pro-actively propose fitting labels and coherent text passages. The thesis lives in the intersection of Recommender Systems and Natural Language Processing. For evaluation, the candidate(s) will conduct a user study with volunteers to measure the perceived ease of use.
The Norwegian Parliament has published a data set of transcribed speeches (https://www.nb.no/sprakbanken/en/resource-catalogue/oai-nb-no-sbr-58/) that the candidates can use for testing purposes. We expect to gain access to a larger corpus from Norwegian news media which can serve as the evalaution set and support the user study.
Texts represent challenges for machine learning methods that require numerical input. Thus, the candidate(s) will ahve to first transfer the text into a numerical form. Knowledge about machine learning and natural language processing is a good asset.
VinDR is an X-ray dataset that contains chest xrays depicting various medical anomalies within the lungs. They are categorized into fourteen abnormal conditions. Classifying between these conditions is a challenging task. This project aims to develop a model that can detect the type of anomaly and localize it.
The aim can be broken down into two objectives/tasks:
The student is expected to test various deep learning models to achieve the above two objectives.
Skills required: python programming—basic knowledge of machine learning, CNN models, and deep learning.
This project is about anomaly detection in time series data, i.e., detecting if a data-point or a short sequence of data-points inside a data-stream should be considered as “abnormal”. Anomaly detection is interesting in many real-world applications, like monitoring machinery to ensure it is a smooth process, finding fraudulent money transactions, etc.
Due to its importance in practice, many different approaches for anomaly detection exist. Hence, the first task in this project is to get a good understanding of the state of the art. Next, we will look at if and how new foundational models for time-series data (e.g., Lag-Llama and TimesFM) can be used for anomaly detection. Our goal is to utilize the zero and few-shot capabilities of these models to make a zero or few-shot anomaly detection system. The resulting system will be tested on real-life data.
If you are interested in this project, please make sure to look at this page.
The object of the current work is to apply machine learning techniques to identify the high temperature softening and melting properties of materials based on images/videos recorded from the sessile drop equipment. The image soft-sensor, developed by this work, will improve data pre-processing and digitalization of the SINTEF wetting lab.
Background:
SINTEF wetting lab focuses on visualizing millimetre-scale samples in sessile drop furnaces. High-resolution pictures will be captured in real-time at high temperatures to evaluate the wetting, softening, melting, reduction, gas production, foaming, and dissolution kinetics, etc. Evaluating thousands or millions of pictures is time – consuming and deviation is verified with various operators. It is challenging to conclude accurate results in an efficient way. In this respect, the use of machine learning is needed to improve the evaluation of the result.
Scope of work:
Skills required:
Data pre-processing skills for image/video using deep neural networks. Super-Resolution technique may be necessary in order to increase the resolution of images.
Reporting:
A written time schedule should be presented at project meetings if necessary. The work should be presented in written form, and common rules of reporting technical work should be met.
Supervisor at NTNU: Prof. Zhirong Yang
Co-supervisor at SINTEF Industry: Dr. Sarina Bao, Dr. Kai Tang
Rather than HPC enabling AI, what can AI do for HPC?
ML has already been shown to be useful when dealing with lots of parameters when optimizing code, but how do we “co-pilot” paralle programming?
Can "co-iploting also be used for eductainment targening Parallel Programming?
These and other related topics for the motivated student(s) that want to delv into how AI and LLMs can bring the parallel programming world forward are welcome to contact Anne Elster for a chat at elster@ntnu.no
This topic is broad, but we expect the topics will be narrowed after discussions and the student´s interest. The fall project will include a literature survey
This project is in collaboration with Trondheim Kommune and the student(s) will get a co-supervisor. The project also involves using the Norwegian large language model (NorLLM) that is developed at NTNU.
Problem Description:
Trondheim kommune is a large organisation with approximately 220 different services to improve the lives of the inhabitants. The services span multiple domains and we have data “from birth to the grave” for all inhabitants. However, utilising this data is bound by many restrictions as by legislation.
We have created a knowledge graph based on RDF/OWL that uses service based codes (“k-koder”) as an identifier for processes associated with it. These processes and data products are also (partially) semantically connected to the organisational structure, entailing over 200 business units/departments spread over 7 main domains. We also have semantically connected some of the processes to our enterprise architecture overview.
All of our services have a link to relevant legislation, as some services are also semantically related to our internal quality system (“kvaliteket”). OntoText describes Semantic Technology as follows: “Semantic Technology uses formal semantics to give meaning to the disparate data that surrounds us. Together with Linked Data technology, it builds relationships between data in various formats and sources, from one string to another, helping create context. Interlinked in this way, these pieces of raw data form a giant web of data or a knowledge graph, which connects a vast amount of descriptions of entities and concepts of general importance.
Semantic Technology defines and links data on the Web (or within an enterprise) by developing languages to express rich, self-describing interrelations of data in a form that machines can process. Thus, machines are not only able to process long strings of characters and index tons of data. They are also able to store, manage and retrieveinformation based on meaning and logical relationships. So, semantics adds another layer to the Web and is able to show related facts instead of just matching words.” (More details can be found from the source:https://www.ontotext.com/knowledgehub/fundamentals/semantic-technology)
We lack proper functionality to:
A) see whether we have duplicate services and/or service descriptions.B) see whether we have duplicate quality articles pertaining to these services.C) search the knowledge graph using the Norwegian language as a prompt
Our hypothesis is that all these points can be addressed by using NorLLM as a means to enrich the graph in order to expand the search capabilities by also searching for, for example synonyms and/or hypernyms. NorLLM has 3 different language models and we would like to investigate which of these 3 works best in terms of said information retrieval tasks.
We expect the student(s) to
● Do a literature review on language processing and deduplication techniquescapable of utilising semantic descriptions.
● Research semantic deduplication techniques when using enriched data by applying expansion using different versions of NorLLM. ○ Evaluate and classify the best performing NorLLM model ○ Evaluate how the best performing NorLLM compares to at least one LLMthat was not specifically trained for the Norwegian language.
● Identify and implement the most suitable manner to query the data in the graphusing the Norwegian language
Dataset:
Trondheim kommune will provide an OWL/RDF knowledge graph describing:● a set of data-driven problems defined in the internal project “Datadrevet Organisasjon” as well as through a series of workshops.
● relationships between the data-driven problems and the “K-koder”, a standard for classifying and describing public sector services and functions.
The knowledge graph will have an underlying ontology describing:● semantics associated with how a data-driven problem is composed of different categories of descriptions (core problem formulation, defined actions to solve the problem, data products illuminating the problem space and relevant datasets) according to the ODA-method developed by NESTA.
● semantics describing how mandatory public sector services should be provided through specific functions (and sub-functions) and how they relate to Norwegian laws and regulations.
In addition, we currently have available a large amount of documents (in PDF format) from “Innsynsportalen” that can be used to frame the enrichment done by the NorLLM models. Note that (currently) much of this data is formatted as PDF, so it needs to be converted into a machine-readable language.
This project will investigate modeling techniques for describing security aspects at the software architecture level, with emphasis on evolving systems and security assessment.
TL;DR: There exist methods for reliability evaluation of systems based on their architecture, for example Fault Trees. We want to investigate how those methods can be applied to machine learning pipelines. This would allow comparing different architectures based on their tolerance to faults.
Project Description
A model is an abstraction of a system that highlights important features of it for a certain purpose, while abstracting the details that are not relevant. Models are used in different ways in software and systems engineering. In particular, they are often used to document the architecture of a system, for example by using UML or SysML diagrams. Other models are used for evaluating quality properties like reliability, safety, or security. As an example, Fault Trees (FT) [1] are a very common kind of model that is used to estimate the reliability of a system.
Constructing reliability models like Fault Trees is typically a complex task, so there has been a lot of work in the literature to design methods to automatically derive fault trees and other dependability [2] models from models of a system or software architecture [3] [4]. The idea is that, from the information documented in the architecture, a lot can be said about how faults can be generated and how errors propagate. A lot of manual modeling effort can therefore be saved. For example, if we know that two components are connected in a client-server pattern, we know that a failure of the server will propagate to the client.
While those techniques are now established for traditional systems, very few works have addressed reliability models for machine learning pipelines. The objective of this project is to understand what has been introduced so far in the literature, and to understand how existing techniques for traditional systems can be adapted to the machine learning context. It is possible to start from existing research on the derivation of reliability models from UML diagrams, or to define something personalized, based on the expertise of the student and discussion with the supervisor. The idea is to start from some diagram of the architecture of a machine learning pipeline, such as the one that can be obtained with TensorBoard [5] or with Netron [6].
This work proposal involves:
performing a literature review on the derivation of reliability models from machine learning pipelines;
define/extend a method for deriving reliability models from a diagram of a machine learning pipeline. The diagram can for example be extracted with tools such as Netron [6].
depending on the work identified in the literature review, the step above will be adapted to a task that can be developed in the timeframe of a semester, in agreement with the student.
The long-term research objective linked to this activity is to define methods that enable system architects to evaluate and compare machine learning architectures with respect to their ability to tolerate faults.
Needed skills and other conditions
Mandatory
Useful to have (optional)
References
In such a thesis, we will focus on one of the various AI solutions based on the biometric sensors (eye-tracking, heart rate, EEG) to enable better learing experiences for students. We will also focus on collecting data from students such as eye-tracking, EEG, heart rate, skin temperature, and facial expressions. These data sources provide information about the students from the different points of view and combining the provide better predictions of students' behaviour and their performance.
Requirements: The ideal candidate will have a background in system design and basic machine learning. Solid programming skills and an interest in hands-on development and experimentation is also a requirement.Programming skills: Python/Java.
For details about the different options please contact kshitij.sharma@ntnu.no
This projects focuses on artificial intelligence (AI) and machine learning (ML) and its application in geophysics, typically also involving the use of high-performance computers (HPC). Beyond the technical AI/ML/HPC and application dimensions, one or more of the factors safety, explainability and sustainability may be important, depending on the joint interest of the student(s) and the professor(s).
For geophysical data, the increased existence of IoT sensors, including Distributed Acoustics Sensing (DAS) sensors, has great potential. At the same time, there is broad interest in many areas of society in better understanding the connection between such data and sustainability, including land-slides, CO2 storage, and so forth.
This project is connected to NTNU's Center for Geophysical Forecasting (CGF), please check here: https://www.ntnu.edu/cgf. Due to CGF, there is much data, especially Distributed Acoustics Sensing data, available for analysis. There is a need for students with AI and ML competence to develop new algorithms for processing and analyzing such massive data sets.
Further, we have at CMU, NTNU, and USGS made progress on predicting air quality and soil moisture by means of various time-series and machine learning techniques. We have developed models that are rooted in deterministic, physically based hydrology, and we study their capabilities in forecasting soil moisture over time periods longer than a few hours. Learned model parameters represent the physically based unsaturated hydrological redistribution processes of gravity and suction. We have validated these models using soil moisture and rainfall time series data collected from a steep gradient, post-wildfire site in southern California [1,2].
In this proposed project, students can build on and extend the efforts sketched above, according to their own interests and skills. The project may for example focus on studying and testing different ML models on challenging IoT (including DAS) data. Typically, the emphasis will be on forecasting in a complex spatio-temporal multi-variate setting, where IoT senors will be sensors that measure geophysical and related data. It is desirable to obtain a certain degree of spatial and seasonal generalizability as well as explainability and understandability. As an example, parameters of our soil moisture model explain various soil properties such as drying and wetting rate [1,2]. Visualizing data, forecasts, and the behavior of different machine learning methods can also be of interest in this project. Students with skills in high-performance computing and interest in handling big data will also be very welcome in selecting this project.
This thesis project will be conducted as part of a larger research project, CGF, where there will be an opportunity for student(s) to interact with a team of experienced researchers.
[1] Aniruddha Basak, Ole J. Mengshoel, Chinmay Kulkarni, Kevin Schmidt, Prathi Shastry, and Rao Rapeta. 2017. Optimizing the decomposition of time series using evolutionary algorithms: soil moisture analytics. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO ’17).
[2] Basak, Aniruddha, Kevin M. Schmidt, and Ole Jakob Mengshoel. "From data to interpretable models: machine learning for soil moisture forecasting." International Journal of Data Science and Analytics 15.1 (2023): 9-32.
Notes:
In this project, the joint interests of (potential) sponsor, advisor and student(s) come together. Typically, the project will be based on a problem described by a sponsor AND previous research by Prof. Ole Jakob Mengshoel AND interests of students. If only one or two of these are present, there is no basis for a project.
+++++
Please send email(s) to (potential) advisor(s) and/or sponsor(s) if you're interested in this project.
Ole Jakob Mengshoel is more or less following Keith Downing's selection process for master students - please read this:
https://folk.idi.ntnu.no/keithd/ai-masters/kd-masters-selection.html
The focus of this thesis is to use multiple sensors and artificial intelligence to predict the various performance measurements of pair programmers. Pair programmers usually produce better programming results than the individual programming therefore it is important to understand the factors that contribute to their success. With multiple data sources providing us with information from a diverse points of view, recent works have shown their advantages over individual data sources. In this thesis, the students will use eye-tracking, EEG, heart rate and facial expressions as the data sources.
In the first step, the students will gather data from multiple wearable and pervasive data sources while the participants are pair programming. In the second step the students will develop prediction algorithm using the features from the interaction of the two programmers. Finally, the students will compare the various algorithms and feature sets and write-up the thesis.
Requirements:The ideal candidate will have basic background in machine learning and deep learning algorithms. Programming skills: Python.
The last few years have seen an explosion in interest regarding the use of Artificial Intelligence and much talk about the potential business value. Nevertheless, there is significantly less talk about the challenge's organizations will face when implementing such solutions and how they should overcome these obstacles. Inhibiting factors are not only of a technological nature but also include organizational and human factors. This project will involve collecting and analyzing data in collaboration with the researchers from the Big Data Observatory (https://www.observatory.no).
Advanced forms of analytics and aritificlai intelligence are becoming increasingly deployed to support the work of healthcare workers. Medical doctors, nurses, and administrative staff either use, or are aided by sophisticated technologies which are posed to radically change the nature of their work. For example, radiologists now rely increasingly more on machine learning techniques to and other applications of AI to diagnose patients, while a lot of procedural and repeptive tasks are being done by machines. The objective of this project is to understand how the nature of work for health practitioners is changing, and what positive and negative consequences they experience.
Description: This project is about artificial intelligence (AI), cognitive science, and computer games. In particular, the goal of the project is to improve the understanding of cognitive functions of humans by having them play computer games, measure their brain activity, and analyze their performance by means of artificial intelligence techniques.
The type of computer game considered in this project is a two-agent shooter game that is relatively short in duration, typically around 2-3 minutes. The specific game proposed, SOL, has been developed, with AI components, in a previous master’s thesis in Mengshoel’s group at NTNU. We foresee in this proposed project a setup with two agents, in which a human subject plays against a computer agent. The computer agent may be developed manually or by means of AI techniques.
Functional magnetic resonance imaging (fMRI) is a technique to measure brain activity by measuring changes in blood oxygenation in the brain associated with task performance. One can thus use fMRI to measure, in real time, how humans respond to varying sensory stimuli and solve tasks, for example in computer games. A subject can play the computer game, while being placed in an fMRI machine, such that the brain activity is also measured. This can, in turn, lead to improved understanding of how cognitive function develops in humans, both in the short and long term.
One potential role of AI methods is to provide a decomposition of the gameplay and human behavior of an analysis algorithm. The fact that a game is multidimensional and not something one has to do all the time, makes it realistic and thus interesting from the perspective of measuring brain activity and cognitive function.
These are a few ideas of specific topics and research questions that can be investigated in this project, depending on the interest and background of the student(s):
1) Personalized feedback to the human player can be provided, based on AI analysis of their gaming performance. How does such feedback impact their later performance? Is there a difference between subjects as far as their capability of productively using feedback and improving their performance over time?
2) One can vary the computer agents and their behaviors to better understand the visual cognition of the different human subjects as far as the impact on their gaming performance is concerned. As an example, one can move the computer agent quickly or slowly during gameplay to study the reaction time of the human subject. What is the “breakpoint“ of the subject? In other words, where does performance of the human subject start to deteriorate as the difficulty and challenge of the game is increased?
3) In addition to having a subject play the computer game only, they can also be taking more traditional cognitive tests such as those previously designed by Håberg’s group and available on the Web. Based on the subject’s performance on the computer game, can we predict the subject’s performance on more traditional cognitive tests?
4) AI methods, for example neural networks used for computer vision tasks including classification, have been found to be brittle and easily tricked by simple adversarial attacks that change visual stimuli in extremely minor ways from a human point of view. Human performance is, in contrast, typically quite robust. Is there a way to study this difference in the game setting, by setting up experiments that compare how human subjects react to a computer game to how AI agents react?
"In the wild" eye tracking is an umbrella term for recording eye movements of a subject that is in movement, opposed to more traditional eye tracking which requires the subject to remain stationary during experiments. This project has the aim of developing a framework for quantifying the (in)accuracy of eye tracking in VR. The thesis will involve finding existing literature in this domain to map existing solutions (if any), and to assess whether the best course of action is to try to improve upon these, or whether developing an entirely new solution seems to be the best course of action.
The results of the last paragraph should lead to the development of a proof of concept implementation where the accuracy of the eye tracking in a given VR headset is quantified, and alongside this, document the results of what works and doesn't work (and why). Ideally, ideas for programatically improving the accuracy of eye tracking in VR will also be developed and tested as a consequence of having developed a better understanding of the mechanics of eye tracking during movement.
This project is connected to research at NTNU’s COMET research group (Computational Magnetic Metamaterials).
The key focus of COMET is to discover new concepts for ultra-low-power computing systems based on emergent physical phenomena. The research targets new neuromorphic computer architectures outside the standard von Neuman type of operation implemented in new materials (beyond silicon and transistors). Of key interest are magnetic metamaterials called Artificial Spin Ice (ASI), consisting of millions of coupled nanomagnets. ASIs have great potential for massively parallel, ultra-low power computing systems.
The project will mainly be based on simulations using our open-source flatspin simulator for large-scale ASI systems.
If this sounds interesting, please get in touch!
Keywords: unconventional computing, artificial life, neuromorphic computing, artificial spin ice, reservoir computing, bio-inspired systems, complex systems, self-organization, emergence, new technology, neural networks, nanotechnology, material computation, nanomagnetism.
Project description
The goal of this project is to explore new clocking protocols to control the dynamic behavior of coupled nanomagnet systems. This is an interdisciplinary project at the intersection between computing and physics.
To perform computations in ASI, it is necessary to reliably control their dynamics, i.e., how nanomagnets change state over time. In COMET, we have recently discovered a class of field protocols called "astroid clocking". These clock protocols unlock a step-wise and gradual evolution of magnetic states within the ASI, and offers unprecedented control of the dynamical process in both time and space. We have demonstrated astroid clocking of one type of ASI, where the method allows emergent domains in the ASI to be gradually grown or reversed in a step-wise manner. As the domain evolution is driven by an external clock field, the dynamical process can be paused and resumed at will. In other words, the level of control far exceeds what is possible with conventional ferromagnetic materials, and what has previously been achieved in magnetic metamaterials.
However, this is only the tip of the iceberg when it comes to astroid clocking protocols. There is a wide range of topics to explore, e.g., new clock protocols, astroid clocking on other types of ASI systems, methods for protocol design, etc.
This project aims to widen our understanding of astroid clocking. While the primary focus of the project is to explore astroid clocking using flatspin simulations, experiments on real nanomagnet arrays may also be possible.
Links and more information
Material computation is computation that exploits physics directly computation.
Neuromorphic computing are computing systems inspired by properties of the brain and neural systems.
Unconventional computing can be defined as non-von Neumann type of computing, often taking inspiration from nature using design principles like emergence/bottom-up instead of the traditional top-down.
Artificial Spin Ice (ASI) are meta material systems consisting of coupled nano magnets arranged on a 2D lattice, whose collective large-scale emergent behavior has attracted considerable interest as computing substrates.
The project focuses on integrating medical imaging data with natural language processing (NLP) techniques to automatically generate medical reports in an unsupervised manner. The project aims to utilize autoencoders, a type of neural network architecture used for learning efficient data representations. In this context, an autoencoder will be employed to learn a compact representation of both medical imaging data and associated information, such as diagnostic annotations or patient information. Similarly, with the knowledge graph, a structured representation will be created that captures relationships and connections between various medical concepts, which can include image features, diagnoses, treatments, and patient details. The project aims to bridge the gap between medical imaging data and NLP by leveraging the encoded knowledge graph to generate coherent and informative medical reports automatically. This involves transforming the structured information into natural language narratives that accurately describe the medical images, findings, and clinical insights.
There are many tasks both for indoor and outdoor which lend themselves to being carried out by (semi)autonomous drones. In this project the aim is to develop a system that allows drones to navigate through a building and carry out specific tasks. The system involves a number of main components that allow: dynamically mapping the environment and navigating it without collisions, planning and scheduling missions, managing a fleet of such devices, self-monitor status to ensure returning to the base / charging station.
Examples of tasks could be: inspecting certain areas and taking a photo for a human operator to later check, taking measurements in different spots in the building such as temperature, humidity, co2 etc
The project involves a study of relevant existing research and literature, design and implementation of a functional prototype and evaluation of the developed prototype at different levels, including user testing.
While the project can be assigned to a single student it is recommended that a pair of students will work on it.
This project will build on a previous prototype
There are many tasks both for indoor and outdoor which lend themselves to being carried out by (semi)autonomous robots. In this project the aim is to develop a system that allows rover like robots to navigate through a building and carry out specific tasks. The system involves a number of main components that allow: dynamically mapping the environment and navigating it without collisions, planning and scheduling missions, managing a fleet of such devices, self-monitor status to ensure returning to the base / charging station.
This project will be developed in collaboration with Norsk Helsenett.
The project investigates automated scanning tool for Kubernets, to check whether the Kubernets environments comply with predefined best practices. In particular, the project will develop a customized scanning tool based on the environment in place at Norsk Helsenett. This could include checks of configurations where a range of parameters are assessed, including for example admin rights, RBAC policies, and network policies.
The project will take inspiration from existing tools, such as:
- https://github.com/aquasecurity/kube-bench - https://github.com/aquasecurity/kube-hunter - https://popeyecli.io
This will provide the candidate with learning in Kubernetes, Linux, networking, information security, and software development.
Collaboration
In partnership with University of Stirling, Scotland, United Kingdom
Introduction Salmon skin disease constitute one of the main challenges in the modern aquaculture industry, resulting in substantial economic losses each year. Computational biologists are leveraging new and more accurate methods to better understand the problem. The genomic analysis requires accurate annotation of single-cells in salmon samples, which is currently the most significant bottleneck in the uptake of this technology. The manual annotation process is time-consuming, costly, and error-prone. To address these issues, our project aims to leverage state-of-the-art machine-learning software libraries to develop an automated tool for predicting single-cell types in salmon genomic samples.
Project Objectives
Project Description The project will provide the student with a unique opportunity to delve into the world of aquaculture and single-cell data analysis. By harnessing modern machine-learning techniques, the student will contribute to the development of a tool that can significantly expedite the single-cell type annotation process, ultimately aiding in the prevention of salmon skin disease.
Skills and Requirements
Benefits for the Student This project offers a dynamic learning experience in which the student will:
In summary, this project presents a valuable opportunity for a motivated student to make a meaningful contribution to the field of aquaculture while acquiring essential skills in machine learning and interdisciplinary collaboration. We invite dedicated and enthusiastic students to join us in this exciting venture.
Background
Language technology is a central application and innovation area in artificial intelligence (AI) and machine learning (ML). It is an enabling technology with a wide range of applications. For example, automatic speech recognition (ASR) has been an area which has contributed significantly to the general development of both AI and ML.
In spite of impressive progress in recent years, conversational speech and colloquial dialect speech pose major problems for speech technology in general, and for the Norwegian language in particular. There are still significant application areas that require innovative solutions and breakthroughs in order to be fully realized. Issues that must be resolved include both language technology in general and applications for the Norwegian language (and other resource-constrained languages). In fact, the importance of improving language technology, including speech recognition, in the handling of Norwegian has been highlighted by the Norwegian government's AI strategy document [1].
Problem Description
This project is part of a language technology initiative involving Telenor, NRK, National Library of Norway (NLN), and NTNU. The main goal of the initiative, hosted at the Norwegian Open AI Lab, is to improve language technology for the Norwegian language, by focusing on (i) speech-to-text transcription system capable of better transcribing multi-party conversations in realistic recording conditions (SCRIBE) and (ii) making text transcriptions (and thereby also the underlying speech content) available via improved automated metadata generation.
In this project, MEGAS, the focus is on the automated metadata generation, including how metadata generated from text data (including transcribed speech) can improve access to, analysis of, utilization of, and overall benefit of the data. In a complementary project, SCRIBE, the goal is to study and develop solutions for machine transcription of Norwegian conversational speech. These two projects are, in other words, mutually beneficial to each other but at the same time somewhat independent.
Data Availability
We discuss two potential data sources below:
(1) This project may use NRK-provided TV-content with subtitle files; we will simply call it NRK’s subtitled TV (NST) dataset. The NST dataset, with 14,614 entities, consists of TV programs with Norwegian speech along with subtitles. This NST dataset will be interesting in its own right and also a stepping stone towards ASR-generated transcripts. For this proposed MS Thesis project, in a language technology domain that is already relatively challenging, we believe that starting from relatively clean text data is quite reasonable. In this project, the main focus will be on automated metadata generation from the subtitles in the NST dataset.
Note that we are proposing to use a «simpler» dataset compared to what one may obtain from ASR-transcriptions from Telenor, NRK, or NLN. Consequently, the project will have a focus on the text processing aspects of language technology rather than «error-correcting» conversational speech transcripts generated using ASR. Still, there are many challenges to address and results from the proposed project would be useful in the more general and complex setting of ASR transcripts of conversational speech encountered by Telenor, NRK, or NLN.
(2) This project may use ParlaMint-NO data provided by the Norwegian National Library. ParlaMint-NO contains the Norwegian part of the ParlaMint project, an EU-funded project supported by CLARIN ERIC. The project’s aim is to create comparable and similarly annotated corpora of parliamentary meeting minutes.
This corpus contains minutes from the Norwegian Storting for the period October 1998 – May 2022). The most recent version of the Norwegian data does not appear to be available on the CLARIN Web page yet, but is available her: https://www.nb.no/sprakbanken/ressurskatalog/oai-nb-no-sbr-77/
There are older and other ParlaMint datasets available from the Norwegian Storting available in CLARIN and elseqhere. The above-mentioned ParlaMint-NO is the most up-to-date, and it has metadata in a well-described, common European format. Thus we recommend using it.
Project and Thesis Descriptions
High-quality language technology for Norwegian will strengthen the Norwegian language, enable efficient digitalization and information retrieval, simplify interfaces to public services for citizens, and provide invaluable assistance for people with special needs, to name but a few application areas of impact. Good quality metadata in addition to transcripts is required in order to enable such applications. For Norwegian, the capabilities of automated creation and utilization of metadata need to be strengthened, and this will – at a high level – be the the goal of this project.
The project can be suitable for one or two students, and the following tasks have been identified.
Sketch for the Project Report:
Sketch for a follow-up MS Thesis, using the dataset(s):
Business or Scientific Restrictions
We foresee no business or scientific restrictions for the project. The candidate will work with a relevant dataset, such as dataset (1) or (2) discussed above. Computational resources are made available by NTNU.
The methods and technologies to be developed will, in addition to bringing groundbreaking impact for Norwegian speech technology, contribute to moving the research front for ASR and ML in general, specifically in relation to other small or under-resourced languages. In addition, the project will build vital competence among the project partners. Project outcomes will be openly and broadly disseminated.
[1] https://www.regjeringen.no/en/dokumenter/nasjonal-strategi-for-kunstig-intelligens/id2685594/
Please send email(s) to (potential) advisor and/or sponsor if you're interested in this project.
The future of Autonomous Driving (AD) is end-to-end approaches (IL, IL+RL), simulators, closing the sim to real gap using Neural Rendering/Reconstruction, TeleDrive over 5G+ if the AD-car need assistance, HD-maps / Digital Twins (DTs) of the infrastructure where you want to drive autonomously, especially in Nordic winter environments (these HD-maps/DTs can also be used in the simulators), crowdsourcing to have an updated Digital Road Twin (e.g. mobile device with accurate positioning and geo-referencing), and automated anonymization of camera data - executed in a way that make the filtered images useful for training and verification of AI models.
Within NAP-lab (= Ntnu Autonomous Perception lab), and associated projects (e.g. MCSINC, MoST etc.), we are working on all these topics (at a state-of-the-art level).
Among other things, our research platform consists of a KIA e-Niro, equipped with NVIDIA DRIVE AGX Orin compute and software stack, 8 cameras, 3 LiDARs, 2 radars, accurate GNSS with CPos corrections, and last but not least, a CAN-bus based drive-by-wire module (DriveKit) that makes it possible to control steering (lateral control) and throttle / brake (longitudinal control) of the car using an API (see the first part of this video). Our old KIA will also be equipped with comma three / openpilot for low-tech end-to-end AD research before the semester starts. On the simulator side we have been using CARLA for quite some time now and will be among the first academic institutions that will have access to NVIDIA DRIVE Sim. In terms of HD-map / Digital Road Twin development we are using NVIDIA Omniverse.
With your help we would like to continue our AD-work within the following areas:
End-to-end Autonomous Driving (but also the more traditional Modular approaches), first in a simulated environment and then test the system in real life using our KIA AD research platform: E2E IL EarlyTest, more info 1, more info 2
Neural Reconstruction (to close the sim-to-real gap): NeRF CARLA EarlyTest, more info
TeleDrive (first and foremost as a backups-system when an AD assistance is needed): TeleDrive KIA EarlyTest
HD-maps / Digital Twins (DTs): Digital Road Twin (from one of our test areas)
Control algorithms for autonomous driving (Sim & Real). More info.
Crowdsourcing (mobile app with accurate geo-referencing for updating the HD-maps++). More info.
Anonymization of camera data: DeepPrivacy2, more info
These are just some examples, many other topics exist, and you are welcome to come up with your own project within the AD domain.
For many of the proposed projects, an AI / Computer Vision (CV) background is of course beneficial, but there are also projects where students with other specializations are highly welcome.
We encourage and help to publish papers based on some of the master-thesis work done. For those that are up to it it's also possible to have a research assistant position on the side. It migh also be possible to proceed as a PhD candidate within AD after you finish your master, or as an integrated PhD right before you start on your last master year.
You can work individually, or in pairs of two.
Main investigators for NAP-lab related project are Frank Lindseth and Gabriel Kiss
The start-up company Zeabuz, based in Trondheim, works on the development of autonomous passenger ferries. Such systems are strongly dependent on reliable and powerful AI methods for environment perception ("where am I?", "where are other ships, persons, piers, ...?") and for situation assessment.
In cooperation with Zeabuz, there are several student projects (specialization projects, Masters theses) such as:
* AI-based motion planning for autonomous ferries
* Stereo vision for autonomous ferries
* Object detection in visual and infrared imagery under adverse weather conditions
* Automatic labeling of infrared images
* Reinforcement Learning for Black-box Safety Validation of Autonomous Marine Vessels
* Wake simulation for testing of situational awareness
Link to the complete Zeabuz' list of student projects:
https://zeabuz.com/students/#proposals
These projects are to be carried out in direct cooperation with Zeabuz, while the students will be linked to the Environment Perception Group lead by Prof. Mester and directly supervised/advised by Mester and colleagues.
Students who are interested in one or several of the topics above are recommended to contact Rudolf Mester directly. (Signing up in the IDI topic list only is not sufficient).
The project proposes an innovative method for anomaly detection in medical data by combining neural networks with symbolic reasoning techniques. The project explores neural symbolic approach which refers to a hybrid methodology that combines neural networks, which excel at learning patterns from large amounts of data, with symbolic reasoning, which focuses on representing knowledge and making logical inferences. Neural networks can learn intricate patterns from raw data, while symbolic reasoning can help in explaining the detected anomalies and providing contextual understanding
Computational generation of video game content, often referred to as procedural content generation (PCG), holds much promise for generating character mechanics. Character mechanics refers to how characters are allowed to move and behave in a computer game, rather than aesthetics such as graphics and audio. This projects builds on existing work in which we have studied how to generate character mechanics automatically, by means of novelty search [1,2]. Our results show that some of the auto-generated characters are, by human subjects, perceived as more interesting than built-in game characters.
The goal is to create an AI system for generating fit characters for a simulated environment, a computer game. The focus will be on a game environment, and a key takeaway of the project will be how the methods applied may be used in other applications and other simulations.
The game environment is a simple action game called SOL, and the generated content will be characters in the game. The fitness of the generated characters will be based on how interesting and fun they are to play.
The method to accomplish this task will be based on generation and evaluation of characters by means of evolutionary algorithms and bio-inspired AI. Generation by search and evaluation by simulation will be researched. A variant of adaptive stress testing will be tested for this purpose.
The project builds on an NTNU master's thesis with an accompanying Github repository. Proposed areas of work for a project and MS thesis include these: (1) study alternative diversity preservation GAs (alternatives to novelty search); (2) formulate the various goals of game characters using multiple objectives and multi-objective GAs; and (3) develop a mixed-initiative approach in which a human designer uses SolEA as a source of inspiration but is also involved in the process of game character development.
[1] E. H. Skjærseth and H. Vinje. 2020. Evolutionary algorithms for generating interesting fighting game character mechanics. Master’s thesis. NTNU, Trondheim, Norway. https://hdl.handle.net/11250/2689497
[2] E. H. Skjærseth, H. Vinje, and O. J. Mengshoel. Novelty Search for Evolving Interesting Character Mechanics for a Two-Player Video Game. The Genetic and Evolutionary Computation Conference, July 10-14, 2021.
Although large-language models (LLMs) based on transformers attract most of the attention these days when it comes to neural networks, these systems require an excess of computational resources and produce results without clear explanations: they are essentially black boxes.
For years, some of the most influential deep learning researchers (such as Geoffrey Hinton and Yoshua Bengio) have stressed the importance of going back to biology to find more useful principles on which to base our AI systems. There is already a huge literature on NN models that come much closer to neuroscientific theories and findings. One of the most recent, and more successful such model is the Liquid Neural Network, developed at MIT, based on the C. elegans worm's neural circuitry and containing only 19 neurons. Impressively, it's performance far exceeds that of conventional deep-learning-based controllers that rely on millions of neurons.
Although very new, the liquid network concept derives from much earlier work by Randall Beer, Jun Tani and others on Continuous Time Recurrent Neural Networks (CTRNNs). The CTRNN always had issues with learning (and was often adapted via evolution instead), but the MIT group seems to have tackled the learning problem via dynamic time constants.
In this project a student or group can investigate any of a wide variety of biologically-inspired models, although approval of the chosen model will need to go through the advisor sometime in the first month of the project, during which the student can investigate different models. The student(s) will then compare their chosen model to a few, more standard, deep-learning models on a small collection of tasks to get some idea of the true utility of the model.
For a collection of papers on this topic, see:
https://folk.idi.ntnu.no/keithd/master-projects/2024/
And for more on having Keith Downing as an advisor:
https://folk.idi.ntnu.no/keithd/ai-masters/kd-masters-advice.html
Our planet is changing, now more than ever before. Understanding these changes and how they impact the environment is crucial for preserving the Earth for the coming generations.
Improvements in remote sensing technology and data collection allows us to harvest more data than ever. Hyperspectral remote sensors gather data about electromagnetic radiation reflected off the Earth in nearly the entire spectrum of light emitted from our sun. This data can enable the classification and monitoring of changes in vegetation, agricultural areas, water contents, human habitation, natural disasters and so much more. However, the high dimensionality and redundancy of this data provides a unique challenge for AI.
In 2018/19 a masters student addressed unsupervised spectral band selection of such hyperspectral data, based on clustering bands in highly correlated subspaces and multi-objective evolutionary search using NSGA-II. Experiments showed promising results on several popular hyperspectral datasets compared to other similar recent methods, indicating that this is an interesting avenue for further research. This year students are applying Particle Swarm Organisation to such spectral band selection, applying superpixel analysis.
The project is open for students interested in continuing one of the existing project or applying either a different bio-inspired technique to remote sensing imaging or focussing on another aspect of the imaging challenge.
Local learning rules are believed to be more biologically plausible than training models through backpropagation, and may thus overcome model susceptibility to, for example, catastrophic forgetting and adversarial attacks. The article https://arxiv.org/abs/2205.00920 introduces a novel learning rule that learns input patterns without supervision, however, further study is needed before the presented method can be used to solve more challenging machine learning tasks.
This project aims to continue the work towards biologically plausible learning methods with the overall goal to train more robust machine learning models. For instance, the aforementioned learning rule was used for online arbitrary shaped clustering in https://arxiv.org/abs/2302.06335.
This thesis uses Biometric data (heart rate, EEG, eye-tracking) to understand how the brain process visual conceptual models. Conceptual models are written in specific diagrammatic languages (two dimensional visual models) such as UML and BPMN. A lot of work has been done on the understanding of how humans comprehend and use such models in information systems and software development from the point of view of IT, cognitive psychology and linguistics. On the other hand, there are limited work on how the brain processes such models. Some work is done in neuro-lingustics, but primarily looking at natural language texts. Several areas of the IT-field has also used techniques from neuro-science for a while (NeuroIS where one look e.g. on the usage of IT systems and appropriate user interfaces, and NeuroSE where one in particular look on comprehension of software code)
The task consist in establishing an overview of current work in neuro-science, neuro-linguistics and IT relevant for understanding how the brain process conceptual models as part of model comprehension, and develop and conduct experiments to investigate aspects of this, including how visual conceptual models are processed under comprehension, if there are individual differences in the comprehension of models based on personal characteristics and if there are certain ways of modeling that are more appropriate than others seen from the point of view of human processing. The report is expected to be written in English, and a good master thesis will be able to contain material for a scientific publication.
Vi ansetter også forskningsassistenter inn mot tematikken . Oppgaven gjøres i samarbeid med Kshitij Sharma ved IDI og andre samarbeidspartnere ved NTNU
Project Goal: Explore LLMs potential in classifying courses' learning objectives into different levels of Bloom's taxonomy.
Description
Bloom's taxonomy of learning objectives classifies learning objectives into six levels of learning, including remember, understand, apply, analyze, evaluate, and create. Usually, action verbs are sufficient to identify the level of a text (maybe a question item or a learning objective). However, in many cases, action verbs overlap; they appear in varying levels. Therefore, classification based on action verbs only may result in incorrect classification. This thesis idea relates to assessing the potential of LLMs in the identification of text's bloom level.
Objectives/Tasks:
Programming language: Python.
Skills required: Machine learning, Deep learning, knowledge about prompt engineering, and LLMs.
Bakgrunn og motivasjon: For å møte klimautfordringer må Europeiske selskap integrere bærekraft i alle aspekter av forretningsdriften. Med innføringen av det nye Bærekraftsdirektivet fra EU (Corporate Sustainability Reporting Directive, CSRD), som trer i kraft i januar 2025, stilles det strengere krav til bedrifters rapportering av bærekraftsdata. Dette inkluderer detaljert rapportering av CO-utslipp, spesielt såkalte scope 3-utslipp, som omhandler indirekte utslipp i verdikjeden.
En viktig grunn til virksomheters utfordringer med bærekraftrapportering er at nøyaktig hva som skal dekkes av “bærekraft” er under-spesifisert, hvilke av virksomhetenes data som er relevant, hvordan data skal sammenstilles og hvem bærekraftrapportering er relevant for og hvordan.
Masteroppgaven utføres i samarbeid med Aneo AS (aneo.com, tidligere TrønderEnergi), et ledende nordisk fornybarselskap.
Målsetning: Dette masterprosjektet vil utforske utfordringer og løsninger knyttet til innsamling, behandling, og rapportering av bærekraftsdata for å møte de nye kravene.
Hovedveileder: Eric Monteiro, IDI/ NTNU
Medveileder: Kathrine Vestues, Aneo AS (kathrine.vestues@aneo.com)
Info om CSRD:
Generative AI has shown to have great potential for many tasks previously reserved for humans. The technology is trained with a large collection of sequences and learns statistical patterns.
Whenever possible harm to humans' health and life is at stake, we need to examine a technology carefully. The SFI Autoship explores ways to automate maritime navigation.
Such navigation is subject to rules such as COLREG (https://www.imo.org/en/OurWork/Safety/Pages/Preventing-Collisions.aspx) for preventing collisions at sea.
The candidate has the task to create a generative AI model that takes sensor data and maritime regulations and learns to suggest the decision that minimizes the risk for collisions while obeying legal regulations. The system ought to give instructions to personnel in natural language.
Sreekant Sreedharan and Børge Rokseth will support this master project.
Headaches affect many people. Today, we have access to large data sets describing patients' symptoms, medical measurements, and personal perception. It remains unclear what causes headaches and what remedy is most promising. The master project will explore the use of Artificial Intelligence, particularly Causal Inference, to estimate the causes of headaches in patient records. Causal Inference relies on a scientific model that captures existing knowledge and plausible assumptions. The candidate will get in touch with medical experts to learn more about those. The goal is to construct a causal graph and estimate the effect of random variables on the observed outcome. The outcome can either be modelled as the probability for observing a headache or the probability that a remedy will reduce the perceived pain.
The candidate should have attended courses in about Artificial Intelligence, Probability Theory, and Causal Inference. Ideally, the candidate has a vested interest in the topic of headache. Besides, the candidate ought to have sufficienct experience in probabilistic programming. Experience with medical data is a plus.
This topic is addressing the problem of characterizing the performance of a detector for marine objects of interest (mostly ships) that works on different sensor modailities (video, radar, Lidar). The detector is characterized on an abstract level, and its performance is expressed by statistical quantities. The core of this project is to provide an abstract representation of such a sensor which can be used in a system simulation context. So there is only little consideration of specific algorithms for video, RADAR, or Lidar sensor data processing; it is only the overall input-output relation inside a complex system which is regarded. However, the multitude of different operation conditions (weather, visibility, ...) makes the problem challenging, as this variability should be represented in the simulated detector.
The project builds on very successful outcomes of a MSc project performed between Aug 2022 and May 2023.
This topic is performed in close cooperation with DNV.
Context: RISC-V is an open-source instruction set architecture (ISA) offering a customizable framework for designing processors. RISC-V promotes the capability to define custom ISA extensions to customize processors for specific applications or performance requirements.Problem: Silicon is fixed in atoms, and so are compiler targets. How can compilers help a designer who is still in the process of defining the capabilities of a new architecture in terms of ISA features? What if we want to explore a set of alternative designs?Goal: We want to make the LLVM compiler a valuable tool in the early stages of hardware design by being able to dynamically define custom compilation targets. For example, if a designer must create the best accelerator for a specific domain (e.g., autonomous driving or post-quantum cryptography), what are the most useful vector instructions to accelerate? Can they automatically recompile the same program for a wide range of variations of the same ISA and assess the impact on performance?Requirements:Programming Languages: C/C++ (mandatory), Python (desired)Tools: CMake, GitOS: *nixCompiler toolchain: LLVM (desired)English language: working proficiency (mandatory)
This project is in collaboration with:
This project will explore the concept of Urban Living Labs as a means of engaging citizens in cites and neighbourhoods to design services to meet the need of the citizens. The aim of the project will be to establish a small scale Living Lab, either at NTNU or Trondheim city, and to experiment with the Living Lab. The focus of the Living Lab could be focussed on several things, depending on the project; for example, health and well-being of citizens (SWELL project: https://www.ntnu.edu/sustainability/swell).The tasks will include:- A literature review of Urban Living Labs for cities- Selection of methodologies for establishing a living lab; e.g. co-creation methods, data evaluation methods, feedback and iteration processes.- Identify a specific focus of the living lab.- Establish a Living Lab.- Validation.
Classifying animals in the wild in complex backgrounds is a challenging and open problem. The complexity increases with natural vegetation, varying environmental conditions, half-captured animal parts in frames, lightning variations, and so on. This project aims to classify not only the animals captured in the frame (i.e., elk, rabbit, deer) but also the background surroundings into snow, grass, trees, etc. The project entails developing deep learning models to label foreground objects (animals) and the background into distinctive categories. Elephant Expedition (EE), Snapshot Wisconsin (SW), Snapshot Serengeti (SS), and Camera catalog (CC) to identify animal species in the wild are some of the datasets that contain millions of images that can be used for training and testing deep learning models. Norwegian wildlife dataset collected by NINA is also available with us.
The project aims to study various aspects of creating interactivity in large classrooms. Potential scenarios could include individual or group work combined with discussions, ad hoc quizzes, practical activities etc.
The project involves a study of relevant existing research and literature, planning and conducting a series of user tests. Furthermore it is expected that such test will generate a wealth of data to be analyzed and interpreted to draw out interesting and useful results and conclusions. Depending on the case it might be necessary that the students develop data processing scripts and or novel visualizations.
Reverse engineering (RE) is the process of discovering features and functionality of a hardware or software system. RE of software is applied where the original source code for a program is missing, proprietary, or otherwise unavailable. Motivation for RE ranges from extending support of legacy software to discovery of security vulnerabilities to creating open source alternatives to proprietary software.
RE usually targets binary programs with a known instruction set architecture (ISA) and executable format. The RE process proceeds by disassembling the binary into assembly code, and where possible decompiling the assembly to yield high-level source code (for example, C source code).
However, in many cases the ISA is either undocumented, unknown, or unavailable. In addition, some types of malware uses custom virtual machines to avoid detection. Such cases prove extremely time intensive for the reverse engineer. ISA features such as word size, instruction format, register size, and number of physical registers are a prerequisite to disassembly.
This project aims to discover to what extent machine learning can be used to detect ISA features from binaries of unknown provenance, and if so, whether these features can be used to help disassemble the binary program so that instruction and control flow information can be recovered.
The hypothesis to be explored in this project is whether ISA features can be deduced from binary programs represented as images input to convolutional neural networks (CNNs). As a starting point, each feature (e.g., word size, instruction width, etc.), could be extracted by a distinct trained CNN.
Useful experience for the project includes good knowledge of computer architecture and assembly, machine learning (using Python), and a passion for staring at random-looking byte sequences for hours at a time.
How can we help students to learn more about the negative and positive impact of technology on sustainability? This task will focus on the development of collaborative games for increasing awareness about the role of IT in reaching the UN Sustainable Goals. Students are welcome to discuss specific areas of interest, both with respect to specific Sustainability Goals that they want to address, as well as game genre and technology.
Previous work has been done in the group about teaching about sustainability to computer science students with games and provides a good starting point, still giving freedom to shape your work.
The task is expected to include design, prototyping and evaluation.
Contact the supervisor to share your ideas and know more about this task
Electroencephalography (EEG) is an electrophysiological monitoring method to measure electrical activity in the brain. From noninvasive small electrodes that are placed along the scalp, EEG record spontaneous electrical activity in our brain. Analyzing EEG signal data helps researchers to understand the cognitive process such as human emotions, perceptions, attentions and various behavioral processes.
Optical motion capture systems allow precise recordings of participants’ motor behavior inside small or larger laboratories including information on absolute position. Movement tracking is useful in, for example, rehabilitation settings where tracking a person’s movements during training can give valuable information to provide feedback on whether an exercise is performed correctly or not. Traditional systems often use specialized sensors (Kinect, accelerometers, marker-based motion tracking) and are therefore limited in their area of application and usability. With advances in Machine Learning for Human Pose Estimation (HPE), movement tracking has become a viable alternative for motion tracking. Combining HPE-based movement tracking and EEG can provide the patient with more holistic feedback and help with progress in rehabilitation.
In this study, the students will combine EEG and HPE-based movement tracking, for an existing VR exergame. The EEG can be used to give the patient additional feedback on, for example, her/his attention level, movement intention, or cognitive load. For comprehensive analysis of bio-signals tracked by various sensors, movement and brain data need to be time-synchronized with VR contents. The pipelines that could be used are https://www.neuropype.io/ or https://timeflux.io/ based on LabStreamingLayer. They should work with our EEG, Unity out of the box, however, are not yet synchronized with HPE.
The study is associated with Vizlab and the needed sensors and the basic training on how to use them (VR headset, EEG equipment) are provided.
The eye-tracking technology has advanced in the last decades so much so that now there are available eye-tracking based input devices (gaze interaction). The goal of this project is to compare the performance of several combinations that involve eye-tracking input devices with traditional mouse input devices for a number of typical tasks (web browsing, email, report / essay writing, code editing).
When building Large Language Models (LLMs) for Norwegian, acquiring sufficient alignment data is a challenge. This data aims to ensure that the models output aligns with the values and ethical standards of the model creators.
Using synthetic data to countervail data shortages is a normal practice today, e.g., surveyed by Joshi et al. (2024) https://ieeexplore.ieee. org/abstract/document/10423161. However, in a recent report by the RAND corporation about Artificial Intelligence (AI) in the context of war (https://www.rand.org/pubs/research_reports/RRA1722-1.html), it is claimed that “AI cannot make up for a scarcity of appropriate data.”
In this project, the candidate will compare using synthetic and real data for the models created in the TrustLLM and NorLLM projects, both in terms of qualitative and quantiative evaluation. The above problem statement must be operationalized into questions of a form that is possible to answer with scientific experiments of appropriate scope.
Lars Bungum <lars.bungum@ntnu.no> (post.doc.) will be co-supervising the project.
Approximate computing is the science that studies how to provide ‘good enough’ results -- according to an application-specific quality metric -- while at the same time improving a performance metric such as time-to-solution, energy-to-solution, area, etc.
Among many approximate computing techniques, precision tuning focus on modifying the data types used in the computation in order to reduce the number of bits used to represent real numbers, and/or replace floating point data types with fixed point numbers or other numeric standards, depending on the hardware capabilities.
Precision tuning requires a various range code analyses, transformations, and verification tasks, each of which can be subject of small or big project ideas.
This project will involve international collaborations with:
For your own safety, please consider the following expected requirements and you will not hurt yourself while working on this project.
Mandatory:
Desired:
Main contact: Stefano Cherubin
Equation-based modelling and simulations languages offer a high-level interface to allow modellers to define the behaviour of a dynamic physical system in terms of Differential Algebraic Equations (DAEs), a key element in the simulation and digital twin approach to system modelling.
Numeric integration algorithms can be used to obtain a description of the evolution of the physical system over time by observing its key elements (state variables).
The current state of the art allows for automatic translation of declarative models into executable simulations. However, the current translation algorithms do not scale with large number of state variables.
Ongoing research efforts aims at:
Other contacts: Tor Andre Haugdahl
Disclaimer: this project is a clone of “Compilers for Differential Algebraic Equations” because I am curious to figure out which name sounds more interesting for Norwegian students.
Read also: Writing a Master's Thesis in Computational Creativity
To be creative, we need to produce something which is new, meaningful and has some sort of value. Computers are able to support humans in creative processes, but to also themselves be creative or to assess if an idea or a product is creative. A computational creativity project can investigate any creative field matching the interests and backgrounds of the student or students (language, design, music, art, mathematics, computer programming, etc.), and concentrate on one or several aspects of computational creativity, such as the production, understanding or evaluation of creativity, or on computer systems that support human creativity.
In particular, a thesis in computational creativity can investigate the transitions between different creative artforms, e.g., generating music or images based on textual input (as in Stable Diffusion models), generating music based on images or text, or generating text based on music or images.
Creativity can be found in nature and in humans, but also in computers, and entails to produce something which is new. However, just “newness” isn’t a sufficient condition for us to consider an idea to be creative, it also has to have some value and meaning: If a 2 year old draws some lines on a paper, we rarely consider it to be art; while if a grown-up does the same, we interpret it as having some deeper meaning – and if the grown-up signs the paper with a well-known artist name, we attribute both an underlying meaning and a monetary value to it. Creativity is thus something which isn’t only a result of the effort of a producer, but also very much the result of how the result is viewed by the consumer.
Computational creativity can involve computer programmes that in themselves are creative, but also systems that are able to recognise and access creativity, as well as programmes that assist humans in creative tasks. There are many creativity-supporting systems (e.g., Adobe PhotoShop), and a few systems that themselves (possibly) are creative, such as “The Painting Fool” and “AARON” (two artificial artists). There are also systems that draws art based on textual or musical input (such as Stable Diffusion), or generates music based on images, or systems for automatic captioning of images and videos that produce short texts matching the visual content (with a specific challenge if parts of the texts may need to be emphasised). A master thesis on the topic could address any of these strands and approaches, depending on the student(s) background and interests.
There are three basic machine learning paradigms: supervised learning, unsupervised learning, and reinforcement learning. Supervised learning requires data instances annotated with the correct output labels, but the other paradigms assume no such labelled data. Reinforcement learning (RL) is primarily concerned with what actions an intelligent agent (or group of agents) should take in a specified environment in order to maximise some cumulative reward, while simultaneously exploring the environment and exploiting previously accumulated knowledge.
Within computational creativity applications, reinforcement learning from human feedback has been utilised for language generation in systems such as ChatGPT but can tentatively have a wider usage in art or music, or can be used form part of a generative adversarial network (GAN) that so far mainly have been explored in image generation.
Read also: Writing a Master's Thesis in Language Technology or Computational Creativity
Computational linguistic creativity can primarily be aimed at creating systems that either are creative themselves (e.g., generate poetry, generate screen plays, write song lyrics, produce analogies or metaphors) or try to understand creativity (e.g., identify sarcasm, understand humour or interpret rhymes).
A master thesis project in the field could concentrate on one or several of these different aspects of computational linguistic creativity (e.g., generate and evaluate computational poetry, translate on-line jokes between two languages, or generate plot twists in movies).
Computers have been used in music both as support for creativity and as creative agents themselves, and both for the composition of the music scores and for writing lyrics. The first algorithmic composition system appeared already in the 1950s (the Illiac suite, Hiller & Isaacson 1958), and since then rule-based systems, stochastic methods, grammar-based methods, neural networks, and evolutionary methods have all been utilised to compose music in a specific genre, style or mood; to identify which key(s) or moods a music piece is written in; for generating lyrics; for music information retrieval; or for automatic music transcription. A master thesis on the topic could address any of these strands and approaches, depending on the student(s) background and interests.
Key words: unconventional computing, neuromorphic computing, artificial spin ice, reservoir computing, ,bio-inspired systems, complex systems, emergence, new thecnology, neural networks, nanosystems, material computation.
Material computation is computation that exploit computational properties in underlying physics of material.
Neuromorphic computing is computing systems that are designed and operates in a fashion inspired by the brain and neural system.
Unconventional computing can be defind as non-von Neumann type of computing. Often based on bio/natuur-inspired design principles like emergence/bottom-up instead of the traditional top-down.
Artificial spin ice (ASI) are systems of coupled nanomagnets arranged on a 2D lattice. The collective behavior and large-scale emergent behaviorhas have attracted considerable interest as a promissing metamaterial for compiting.
The unconventional computing and material computing research group at NTNU has shown that ASI is a promissing computational substrates within the reservoiir computing framework. At present time the results are general, basic computational properties. To further demonstrate capability this project will focus on exploring the possibility to adapt the theretical findings to defind real world tasks.
The project will mainly be based on simulations using the flatspin simulator, flatspin is am open-source large-scale artificial spin ice simulator, to simulate ASI reservoirs. Two realworld problems are proposed;
The work may include; finding an efficiant input encoding, defining ASI reservoits with required computatuinal power and defining readout strategies. .
The project is closely connected to the research projects SOCRATES and SpinENGIN where nanomagnets ensembles are explored as a new substrate for computation. We uses nanomagnet ensembles that are Artificial Spin ice. We use Reservoir Computation as computational framework to explore our nanosystems. Within the two projects we produce physical samples that are experimentally tested in synchrotrons and in simulations, mostly simulations use the inhouse simulator flatspin.
Reading material:
flatspin: A Large-Scale Artificial Spin Ice Simulator: A paper presenting our flatspin simulator that most likely will be used (running experiments on the IDUN supercomputer). flatspin simulator and quick start.
Computation in artificial spin ice: A paper illustrating measurements of computational properties in Artificial Spin Ice
Reservoir Computing in Artificial Spin Ice: A paper where Artificial Spin Ice is explored within the reservoir computing framework
Sorry if my txt is a bit hard to get. Anyway, if you are interested in bio-inspired systems, nanosystems, complex systems, recurrent neural networks or just curious on new and existing views on computing please drop me a line and we can arrange a meeting (zoom or hopefully IRL).
This project is a cooperation with a leading industrial partner that is one major player in the field of autonomous driving.
Computer vision for robots, self-driving cars, and also for driver assistance very often assumes very good visibility conditions -- the area is very much dominated by the nice weather conditions that are found when driving e.g. in the Silicon Valley environment.
However, autonomous driving or visual driver assistance in the Nordic Countries needs to deal very often with bad weather conditions: rain or snow, thus wet or snow-covered road surfaces, making vision conditions even worse by the froth or spray caused by vehicles in front of the own one.
This project deals with the exploration what the combination of machine learning and classicial computer vision methods can do in order to detect other vehicles in a distance, road markings, side barriers, and other obstacles as early and as robustly as possible.
One way often being taken in the (sparse) literature about this important topic is to try to improve the quality of the incoming image/video material and subsequently employ 'off-the-shelf' / textbook computer vision algorithms. We do not think that this is the best way to proceed; rather than that, the methods of detecting objects (e.g. in spray, snow, or haze) and the methods for estimating distances and the 3D structure of the environment should consider the afflicted image acquisition process from the very start, and design new algorithms that can cope with the image degradations either by modeling them explicitly (classical approach), or by training learning-based approaches.
The topic of visual environment perception and situation assessment under bad visibility conditions is of course not only of interest in conjunction with cars and road traffic, but equally also for marine applications.
The project can build on an existing database of 'bad weather driving'. If the student(s) working on this project are interested, they can be brought into contact with one of the leading industrial research laboratories for driver assistance and autonomous driving, based in Germany. This may include the opportunity to have a limited guest stay at these labs.
Students who are willing to 'dig deeply' both in the available methods of statistical signal processing as well as in very recent developments and architectures from machine learning / deep learning are invited to contact the prospective advisor in order to explore what this project offers in terms of scientific/engineering challenges and requirements w.r.t. courses and contents that are needed for dealing with this topic.
This project aims at building a mid scale mobile robot using a physical robot that we already have in my group, the LIMO robot by Agilex.ai
https://global.agilex.ai/products/limo
The physical platform (LIMO) is already available at our lab. The project aims to shape this system into an actual robot, interacting with people and/or other robots.
Very probably we aill also procure a 2nd copy of this robot, as it turned out to be versatile in its functionality, and easy to use.
The mobile robot to be developed here focuese on at mid-scale simulation of how mobile robots can interact with each other cooperatively in a dense traffic-like situation, possibly also including interaction with humans. Thus emphasis is not on speed, but on 'intelligence', behaviour, and autonomy.
Note: The final title will include the specific task that the vehicle aims to accomplish
Preliminary work plan: • Specify the goal of the vehicle, e.g., communicating with others to enable collision-free navigation, use of visual and/or positioning information to reach a target as fast as possible, completing domain-specific tasks based on sensor information (for instance, get a package from A to B) • Implement at least one intelligent component, e.g., vision, by using state-of-the-art methods. Possible candidates should be evaluated with respect to their advantages and disadvantages.
Important: the project could also be redefined, in agreement between the supervisor and the student(s) to put most emphasis on the software side of the mobile robot, behaviour, and 'intelligence'.
The 'mobile robot platform' offers a very wide range of possibilities to apply ones creativity for the design and the functionality of the targeted platform. The outcome is not necessarily very 'car-like'; it could also go into the direction of a 'social robot'. The well known R2D2 robot represents nicely the other end of the spectrum of how such a robot could look like.
In case of this project being performed as a Masters project, it would be expected that also at least implementing a part of the robot's intelligent capabilities is included in the agreed project goals. This could be vision, but it could also be on the side of interaction with humans, e.g. some speech or speech understanding capabilities.
The goals to be fixed in the project contract are to a wide degree subject to negotiation with the supervisor, but there will be in any case an engineering challenge in the area of communication, perception, behaviour, or control.
We discuss a potential data source below:
This project may use ParlaMint-NO data provided by the Norwegian National Library. ParlaMint-NO contains the Norwegian part of the ParlaMint project, an EU-funded project supported by CLARIN ERIC. The project’s aim is to create comparable and similarly annotated corpora of parliamentary meeting minutes.
The core idea in this project is to take advantage language models during ASR processing.
Key concepts: natural language processing (NLP), language models, context, CTC decoding, sequence-to-sequence models.
Problem description
In automatic speech recognition (ASR) systems, a transcription is generated when a sound utterance is given to the system. This transcription is the result of a prediction model that selects the most likely sequence of characters according to an acoustic model (AM), during a process called beam search decoding. Typically, one re-scores the n-best candidates based on a separately trained language model (LM), and given the weights of both AM and LM, the system outputs the best candidate.
We assume no prior knowledge of acoustics, signal processing and speech recognition, the main emphasis will be on utilizing language models in improving speech recognition systems.
In this thesis we want to:
This thesis will be conducted as part of a larger research project, Scribe and Megas, integrating the student within a team of experienced researchers.
Main tasks
Data
This project will use open-source data such as the Norwegian Parliamentary Speech Corpus [6] and possibly other curated public datasets where context can be used.
Objective
Improve results from state-of-the-art ASR models by adding broader information beyond the current speech utterance. The aim is to achieve that via a context-aware language model tightly integrated into the prediction or decoding stage of the ASR system.
"A comparison of techniques for language model integration in encoder-decoder speech recognition;" Shubham Toshnival et al.; https://arxiv.org/abs/1807.10857
"BERT attends the conversation: improving low-resource conversation ASR;" Pablo Ortiz and Simen Burud; https://arxiv.org/abs/2110.02267
"Listen, Attend and Spell;" William Chan et al.; https://arxiv.org/abs/1508.01211
"wav2vec 2.0: a framework for self-supervised learning of speech representations;" Alexei Baevski et al.; https://arxiv.org/abs/2006.11477
"Robust speech recognition via large-scale weak supervision;" Alec Radford et al.; https://arxiv.org/abs/2212.04356
"The Norwegian Parliamentary Speech Corpus;" Per Erik Solberg and Pablo Ortiz; https://arxiv.org/abs/2201.10881
To ease the migration of software applications to the cloud, more and more organisations are adopting a Platform-as-a-Service (PaaS) delivery model. A PaaS environment enables software developers to focus on application design rather than cloud infrastructure aspects, aiming to foster more frequent software deliveries. PaaS solutions can either be provided by an external provider (e.g., Red Hat OpenShift or AWS Lambda), or developed and managed in-house. For example, organisations like NAV (nais.io), Equinor (radix.equinor.com), Finn.no (opensource.finntech.no) maintains and offers their own platform as open source to the public.
This project will investigate the adoption and use of these open-source PaaS solutions in IT organisations and how effective they are in supporting developers to move applications into a cloud infrastructure. The project could also investigate the rationale behind using PaaS as a cloud delivery model and to what extent these open-source projects can be beneficial for other organisations [1,2].
[1] Wulf, Frederik; Westner, Markus; and Strahringer, Susanne (2022) "We have a platform, but nobody builds on it – what influences Platform-as-a-Service post-adoption?" International Journal of Information Systems and Project Management: Vol. 10: No. 1, Article 4.
[2] Wulf, F., Westner, M., & Strahringer, S. (2021). Cloud Computing Adoption: A Literature Review on What Is New and What We Still Need to Address. Communications of the Association for Information Systems, 48, https://web.archive.org/web/20210615090458/https://aisel.aisnet.org/cgi/viewcontent.cgi?article=4300&context=cais
Today, we frequently encounter conversational agents/chatbots/virtual assistants when engaging with customer services. These AI system facilitate information access. They provide a way to approach our information needs in the form of a conversation, as we would encounter with human-to-human interaction. Conversational agents can be available 24/7 and help the employees to focus on more difficult requests. The very unique language of bankingi challenges conversational agents. Organizations use special nomenclature and sometimes their own terms or brands.
We are interested in the finanical sector in Norway. The thesis will conduct a literature survey on conversational agents in the financial sector focusing on Norway. We strongly encourage investigating systems that are publicly available. Subsequently, the candidate(s) will implement an information access sytem collecting public information from the Norwegian banking sector. This information can include websites, documents, or multi-media content. The candidate(s) will feed the information into a conversational agent and create a proof of concept. We are particularly intersted in the question how to keep the system up to date.
The data should be publicly available. The candidates will have to crawl the data themselves. We will assist with advice on how to efficiently crawl documents and websites.
Workking with language presetns a few challenges. Many AI systems require numerical input. Thus, texts have to be encoded numerically to be useful. The candidate(s) will have to combine knowledge about Natural Language Processing and Conversational Systems.
You bring problem or method. If it is sufficiently cool and difficult we'll do it.
Sentence simplification is a process by which complex sentences are rephrased into simpler sentences while retaining the original meaning (See Feng et al.: https://arxiv.org/pdf/2302.11957.pdf).
Traditionally, Machine Translation has been done on sentence-aligned parallel corpora, making the unrealistic assumption of a one-to-one sentence correspondence in natural language. Creating MT systems that preserve linguistic variation is a hard problem, not least with regard to preference for sentence lengths. One way of addressing this problem is to look at sentence splitting with LLMs, i.e., to split longer sentences into more units that preserve meaning as a preliminary step in translation.
This project will begin by acquiring the available monolingual sentence simplification datasets and fine-tuning LLMs before moving to the cross- lingual stage. There are some parallel corpora that can be used to extract one-to-many and many-to-one translation items for training and evaluation of the LLMs.
The candidate is expected to use the NorLLM and TrustLLM models for evaluation, as well as benchmarking against the state-of-the-art models. This constraint limits the scope of source and target languages, which can be relaxed if the candidate has a specific language pair in mind.
Are you passionate about machine learning and eager to contribute to foundational research in this dynamic field? We are excited to offer a flexible and ongoing opportunity for Master’s students to work on thesis projects in basic machine learning research.
This opportunity is open to students who:
We will discuss more details and help you choose among the potential topics. Please email me (zhirong.yang@ntnu.no) for an appointment. Preferably, also send the following to me in your email.
Selected students will work closely with our team to identify and refine a research topic that matches their aspirations and our ongoing work.
Data-driven data science is attracting a lot of interest.
However, uptake into organizational practice is lagging significantly behind. Why?
The focus of this project/master is to supplment the possibilities provided by data-driven data science techniques with an empirical understanding of the conditions and circumstances for these techniques to be used in practice for consequential decision-making.
Empirical cases to study data science in practice we will have to discuss. Candidates include: the energy sector, healthcare
An example problem situation is the efforts to enhance the accountability and explainability of algoritms (XAI, explainable AI).
The project/master is likely to be part of the EXAIGON project (https://www.ntnu.edu/exaigon) funded by the Norwegian Research council
This is a very open-ended project inspired by the book, “The Sounds of Life” (Bakker, 2022), which covers many different animals, plants and ecosystems, from bats to whales to trees to coral reefs. In each case, the author mentions that AI (normally deep learning) has been used to interpret sounds, but no details are given. The book provides a massive set of references for eager researchers who want to follow up any of the book's 10 examples.
The interested student(s) will read the book and explore one or more of these 10 paths. This will include tracking down relevant data sources, building machine-learning models, and helping us decipher some of nature's sounds.
This is open to students who have an interest and an idea for a project in Visual Computing.
Students will be chosen based on their performance in the Visual Computing courses and the suitability of their proposal.
TDT4195 and TDT4230.
If you are interested in software and system reliability/safety/security we can agree on a topic of your interest, as long as it is adequate for a Master's project. Topics may range from modeling of reliability/safety/security properties at architecture level, application of formal methods (for example model checking), fault injection, testing, etc. Just drop me a mail at leonardo.montecchi@ntnu.no to start the discussion.
This project aims to develop a real-time monocular depth estimation model for Unmanned Surface Vessels (USVs) calibrated on LIDAR and/or AIS data.
Description and project tasks:
Deep-learning based depth estimation methods have improved significantly in recent years, especially with the recent release of the Depth Anything model. The thesis will develop a deep-learning-based model suitable for embedded platforms (especially, the Jetson Orin NX) for monocular depth estimation.
A suitable starting point for this thesis is to distill the DepthAnything model into a smaller model suitable for real-time inference, for example into a YOLO-based model.The thesis should explore calibrating the depth estimates with LIDAR and/or AIS information collected from the USV. The method should not rely on additional human annotation, and solely use annotations from Large-Vision Models (e.g. Depth Anything), LIDAR scans, AIS information, and possibly other sources.Initial datasets including 360-degree camera data, LIDAR and AIS will be provided.
It is expected to integrate the final system into the Maritime Robotics SeaSight system and test it on the water.
Maritime robotics
The Linguistic Landscape (LL) is defined by Landry and Bourhis (1997) as
...the visibility and salience of languages on public and commercial signs in a given territory or region
Quantitative and qualitative methods are used to interpret the various signals sign instigators send out to sign viewers through LL units. Researchers at
NTNU have already created a prototype for data collection (pictures), and this project will expand on this work. Briefly, the method consists of reading the images with OCR and analyzing them with corpus linguistic tools, e.g., word clouds, and language detection. This work can describe systematic differences in language use between commercial areas and within commercial outlets.
Large Language Models can also be used to analyze these data, e.g., by continuing sentences found in the corpus.
When conducting field linguistics (such as data collection), it is very important to respect owners and employees of shopping malls and individual stores. Thus, this project depends on receiving permission from the malls.
Supervisors: Sofia Papavlasopoulou, Feiran Zhang, Boban Vesin
Place: LCI Lab: https://lci.idi.ntnu.no/, Trondheim
Suitable for one or two students
Emotions can greatly influence the learning process. To better understand the emotional aspects of the children’s learning process of design thinking activities (e.g., empathising with the users, defining the problems, ideating solutions, making and testing a prototype, etc), there is a great need to develop an instrument for self-reporting their emotions. For example, a tool called EmoForm was developed as a retrospective paper-and-pencil-based tool for self-reporting children’s emotional changes over time during this process. The motivation for developing a digital tool that enables children to self-report their emotions and learning is to have a quick, easy, and sustainable approach.
The focus of this thesis is to (1) develop a digital tool to enable children to self-report their emotions and learning during design thinking activities and (2) evaluate the effectiveness of such a tool through a user study that collects data from children in real-life situations (e.g., in a classroom).
Thesis Description
In the first step, the candidate(s) will develop a digital tool (e.g., web-based) to enable children to self-report their emotions and learning, e.g., by extending and digitalising EmoForm. Afterwards, they will use an iterative design process to improve the tool's functionalities and usability. Then, the candidate(s) will conduct a larger user study to evaluate the effectiveness of the tool in the context of design thinking activities. Finally, the candidate(s) will analyse the collected data and write their thesis.
Requirements
The ideal candidate will have a background in interaction design and user experience research. Solid programming skills and an interest in hands-on development and experimentation is also a requirement.
Programming skills: HTML, CSS, JavaScript, Python, Node.js or alike
Expected Project Work Packages (WP)
WP1: Literature study on technologies and tools that facilitate self-reporting emotions and learning.
WP2: Iteratively develops and tests the tool for self-reporting emotions and learning in the context of design thinking activities.
WP3: Conduct a larger user study, collect empirical data, and analyse them.
WP4: Write up the thesis.
Supervisors: Sofia Papavlasopoulou, Boban Vesin and Kshitij SharmaPlace: LCI Lab: https://lci.idi.ntnu.no/, TrondheimSuitable for: One or Two students
IntroductionThis focus of this thesis is the extension of online learning tool to provide learning analytics dashboard (e.g. eye-tracking data, interaction data, etc.) and investigate the learners’ cognitive, affective and physiological state based on the collected data. The aim is to enhance learning experiences for students (preferably younger than 18 years old) in design thinking projects using emerging technologies.
Thesis DescriptionIn a first step, the candidate (s) need to review the literature and familiarize themselves with educational theory, tools and best practices on how these tools can be extended with applications like a dashboard that will capture and then analyze/visualize/generate data of students’ interactions. Part of this task is to decide which data are meaningful for demonstrating students' learning processes while working with each tool and in what type they will be captured and sent for analysis. A user study will follow, to empirically test the proposed system. Finally, the candidate(s) will analyze the collected data and write up the thesis.
Requirements: The candidate should have a background in software design, solid programming skills (preferably Javascript) and an interest in hands-on development and experimentation.
Expected Project Work PackagesWP1: Literature review on relevant tools and activities.WP2: Implement the best practices for capturing and visualizing dataWP3: Iteratively develop and test the system to improve users’ experience and finalize the development. WP4: Conduct a user study, collect empirical data and analyze them.WP5: Write-up the thesis.
In today's digital age, it's crucial to manage important documents like diplomas and licenses securely and efficiently. Traditional methods of handling these documents are either outdated i.e. paper based or fragmented and can pose privacy and security risks. However, with new web 3.0 technology like self-sovereign identity and digital wallets, there's an opportunity to improve how we manage identity documents. This proposal aims to introduce a digital wallet platform that can securely store various identity documents such as academic diplomas, driving licenses, boat licenses, flying licenses, shooting licenses, and so on. By using advanced technology, this platform will make it easier for users to access and manage their documents while ensuring their privacy and security.
The main goal is to create a digital wallet platform (mobile + web) that can safely store and manage a wide range of identity documents. This platform will serve as a centralized place for users to keep their important identities documents, reducing the need for physical copies and/or fragmented storage methods. Additionally, the platform will include additional layers of security features, like encryption and biometric authentication, to protect users' sensitive information.
Furthermore, the envisioned platform will be user-owned, user-friendly, with easy navigation and integration with other digital systems. Users will be able to upload, organize, and share their documents with relevant authorities quickly and securely. The platform will also provide automatic reminders for document renewals, helping users stay compliant with regulations.
Here's a summary of the proposed tasks:
A design system is a collection of documented user interface (UI) elements, visual guidelines, and design principles that other people can use or refer to when designing digital products. Notable examples of design systems are Google’s Material Design and Apple’s Human Interface Guidelines. The main benefits of a design system are (1) improving design consistency across different digital products since it serves as a single source of reference that other people can refer to and (2) reduce the development work since UI designers do not need to design everything from scratch.
NTNU is currently involved in a project called OpenRemote, which aims to create open-source design systems for remotely operated machines. The students can choose to design a design system for one of the following domains:
The students will review scientific literature related to UI designs of the domain of interest and also existing UI designs in the market for the domain of interest. Following the human-centered design approach, the student will design and prototype different types of UI elements that will be made available in the design system for the chosen domain. The student will also need to evaluate the proposed UI elements, to make sure each of them could be understood easily by prospective end users. Throughout your design process, you will also receive support from partners affiliated with the OpenRemote project.
The project will be co-supervised by Dr. Taufik Akbar Sitompul (Department of Design, NTNU)
The urgent need to reduce carbon emissions from maritime activities has highlighted the importance of innovative strategies in interaction design, particularly eco-feedback, which has been effective in nudging car users towards more efficient practices. Despite the acknowledged potential for significant emission reductions on ships through behavioral changes in their operation, there are currently no established standards for eco-feedback within the maritime sector.
This project aims to explore the design of eco-feedback systems for ships, which is part of the OpenZeroproject. The project will focus on the development of new concepts and ideas for providing eco-feedback interfaces to ship operators, aiming to encourage a shift towards more sustainable behaviors.
Through this project, you will have the opportunity to impact the future of sustainable maritime operations, driving a critical industry towards a greener and more sustainable future.
Recent advancements in technology have made it feasible to deploy multiple flying drones that can be controlled in swarms by a single operator. This innovative approach to drone management opens up new possibilities for applications ranging from surveillance to emergency response. In collaboration with scientists from the University of Oslo (UiO) who specialize in drone swarm deployment, this project aims to tackle the unique challenges associated with controlling drone swarms.
This project will delve into the design of control systems for such drone swarms, considering both onsite and remote operation setups. The primary goals will include:
Sails are increasingly viewed as the most significant enhancement to the current international fleet of ships, offering a promising avenue for sustainable energy with minimal infrastructure requirements. With a rising number of vessels under development featuring sail support, and existing vessels being retrofitted with sails, the interest in harnessing wind energy is evident. However, despite substantial investments in this area, the maritime industry lacks established user interface design patterns to operate this new sail-supported vessel category.
In this project, you will propose user interface designs that could support the operation of large, sail-equipped vessels. The process could include the following activities:
This project is part of the OpenZero project, so you will also receive support from partners affiliated with the project, ensuring a comprehensive approach to your design process. This project represents a unique opportunity to contribute to the sustainable evolution of maritime transportation by designing user interfaces that enhance the operability and efficiency of sail-supported vessels. Through your work, you will help shape the future of shipping, making it more eco-friendly.
Read also: Writing a Master's Thesis in Language Technology
It can be vital both for mental health reasons and for security to identify users at risk in social media. That can be users showing signs of mental health issues (depression, suicide, eating disorders, etc.) as well as users being at risk of being radicalised (as extremists, school shooters and other types of terrorists, , etc.), or users being targeted by predators or extremist organisations. This thesis work would experiment with applying various machine learners to social media text. To fully utilise deep learning, a substantial amount of data will need to be gathered, either from previous research or specifically for the task.
Writing secure software is a challenging task and with the landscape of known security vulnerabilities changing almost daily, this task is practically unending. There has been some effort in recent years to use machine learning to automatically scan software repositories during the development lifecycle [1, 2, 3].
The goal of this project is to review the state of the art in using machine learning for automatic detection of security vulnerabilities in source code and if time permits to implement and improve upon an existing method. Due to the complex nature of vulnerability research, the approach can be restricted to a certain class of security vulnerability.
The candidates will build a model that takes source code as input annotates the source code with warnings about potential vulnerabilities. Potential approaches are supervised classification, natural language processing (NLP), or anomaly detection techniques. The models can be trained on open source projects and issue lists, common vulnerabilities and disclosures (CVEs), proof-of-concept code from Exploitdb, etc.
[1] https://sci-hub.se/10.1109/MALTESQUE.2017.7882012 [2] https://resources.github.com/whitepapers/How-GitHub-secures-open-source-software/ [3] https://www.microsoft.com/security/blog/2020/04/16/secure-software-development-lifecycle-machine-learning/
The aim of this research project is twofold: First, the project tries to analyze a large number of job posts in the area of “people analytics” and identify the skills that are mentioned in the job posts. We seek to identify both hard and soft skills. Secondly, the project aims to develop an algorithm that automatically identifies the skills within any job post in the area of people analytics. For this project, databases that hold large numbers of relevant job posts will be used, such as Indeed. A long list of skills from Coursera for the fields of management and humanities that could be used for the identification of skills in the job posts will be provided if needed.
Basic tasks:
Literature review
App design and development (solution)
Assess and evalaute the solution with end users
Conventionally, compilers for imperative languages represent code in static single assignment form organized in basic blocks. The Regionalized Value State Dependence Graph (RVSDG, https://doi.org/10.1145/3391902) is a compiler intermediate representation (IR) developed at NTNU that represents control- and dataflow in one unified representation. The RVSDG is a data-flow centric IR where nodes represent computations, edges represent computational dependencies, and regions capture the hierarchical structure of programs. It represents programs in demand dependence form, implicitly supports structured control flow, and models entire programs within a single IR.
Multi-Level Intermediate Representation (MLIR, https://mlir.llvm.org/) was initially developed by Google and is now run as a subproject of LLVM that provides reusable compiler infrastructure. The aim of MLIR is to make it easier to create intermediate representations at different levels of abstraction, known as dialects in MLIR, and enable conversion between dialects through transforms. The reasoning behind this is that certain optimizations are only possible at higher abstraction levels, necessitating additional IRs, and that there is a lot of unnecessarily repeated infrastructure across compilers. MLIR is an attempt to standardize and unify this infrastructure, increasing interoperability and maintainability. MLRI is growing in popularity and is being used by several machine-learning focused projects as well as for hardware description.
Currently, RVSDG is implemented in jlm (https://github.com/phate/jlm) using a custom framework and an experimental implementation of RVSDG as an MLIR dialect has been created (https://github.com/EECS-NTNU/mlir_rvsdg). This project requires getting familiar with the RVSDG IR and MLIR dialects (regions, operations, etc.), to further develop the RVSDG MLIR Dialect. The work can include how to lower from a more abstract MLIR dialect (e.g., affine) to the RVSDG dialect, implementing optimizations that work on the RVSDG dialect, or implement lowering RVSDG into a version of the Handshake Dialect for the purpose of performing high level synthesis (HLS).
As this project uses cutting edge compiler research tools, a good understanding of compilers and C++ is required. The LLVM infrastructure is commonly used in both commercial and research compilers. This makes this project highly relevant if you are interested in working with compilers.
In 1993, 1998, 2003, 2008, 2013 and 2018 surveys were performed to investigate development and maintenance of IT-systems in Norwegian organizations. Comparable data for important areas where also capture in collaboration with Rambøll IT I praksis investigations. Data collection for a similar survey is carried out yearly since 2015. The assignment will be to analyze the quantitative and qualitative data from recent investigations. Together with a literature review, the survey investigations are expected to give us new knowledge about mechanisms affecting resource utilization related to information systems support in organizations. The report should be written in English and is expected to form the basis for scientific publications
Web3 technologies open new and interesting possibilities in games development! In this project a gaming platform will be developed that allows multiple players to play, learn and/or earn points or digital assets.
Playing activities can be different depending on the chosen game (Student shall propose a game!!) For example
The platform will also encourage users to create games or game contents and reward them!
Mechanisms for exchanging rewarded points with other digital assets shall also be proposed and implemented!
Possible research aspects:
Goals:Our primary goal is to improve the efficiency and user-friendliness of our mobile app by developing and implementing a feature that allows teachers and principals to request and assign substitute teachers directly from their mobile devices.
Tasks:1. Understand the current process of requesting and assigning substitute teachers in schools, and explore how this process can be adapted for a mobile interface.
2. Work as part of a cross-functional team to design a user-friendly mobile feature that facilitates the request and assignment of substitute teachers. This will require knowledge of user interface design principles, programming languages suitable for mobile development, and collaboration with team members from diverse backgrounds.
3. Develop the feature using agile methodologies, including iterative development, continuous integration, and regular feedback from stakeholders.
4. Test the feature to ensure that it functions as expected on different mobile devices, improves the process of assigning substitute teachers, and does not cause any unforeseen issues in the system.5. Implement the new feature in our mobile app, ensuring compatibility with existing modules and data structures.
Learning objectives:In this project, Ingrid and Marius will gain valuable experience in software development, including understanding user needs, mobile app design, programming, testing, and implementation. They will also gain experience working in a cross-functional team, effectively communicating with a variety of stakeholders, and managing a project from start to finish.
The way we write texts give a lot of information about the background personalities of the authors: their age, gender, native language (if writing in a foreign language), if they're human or bots, and possibly their actual identity. This type of information can be used to, e.g., give fair indications of user profiles, to deduce if a text (or a part of it) has been plagiarised, or to uncover social media software misuse. The thesis could thus focus on tasks such as author profiling (what can we say about the author, e.g., their gender, age, if they're a human or a bot), author identification (did a specific author write this text?) and/or plagiarism detection (did somebody else than the author claiming the text actually write all or part of the text?)
I sammenheng med Campus-prosjektet instrumenteres ulike rom med sensorerer for inneklima og tilstedeværelse. Oppgaven går ut på å utforske data fra denne type sensorer i forhold til å finne gode anvendelser av disse.
Prosjektet vil også relatere seg til forskningsrådsprosjektet 'Smart workplaces past Covid-19 som er et samarbeid med Mazemap og CISCO, og arbeid innen FME-ZEN Zero Emission Neighbourhood
Instersted in Digital Twins (DTs), for self-management of health and decision support or road infrastructure?
Have a look here
Stadig mer av oljeutvinningen på norsk sokkel foregår på svært utilgjengelige felt på store havdyp. Samtidig benyttes stadig oftere ubemannete subsea anlegg som ligger på havbunnen men fjernestyres fra land eller nærliggende rigg. Sensorbasert informasjon (bl.a. temperatur, trykk, seismikk, magnetiske og strålingsegenskaper til bergformasjonen) spiller stadig større rolle i alle fasene: leting ("exploration"), boring ("drilling"), logging og produksjon.
Det (nesten) eneste man har er sensorbasert informasjon - men sensorene er upålitelige. De slites (levetid typisk et drøyt år), de er ikke kalibrert osv. Så hvordan skal operatørene ta gode beslutninger gitt høy usikkerhet? Hvilken rolle har nye IT verktøy?
Uansett hva man måtte mene om fossil energi ift klima, digitaliseringen i bransjen er et faglig interessant case for industriell IoT anvendelse med stor overføringsverdi
Oppgaven er tilknyttet NTNUs program for digitalisering i olje, BRU21 http://www.ntnu.edu/bru21
Medveileder: Vidar Hepsø, Equinor
Electric power is a particular type of commodity: it cannot (easily) be stored, needs to be transported over (large) distances for consumption, it is tightly regulated both nationally and within the EU
A central challenge for the electric power markets is the increase in complexity driven by new forms of flexibility, both at the production (eg renewables) and consumption (smart meters, scheduling of consumption based on prices) side. The monitoring, running and planning of these markets are based on comprehensive data platforms that are the basis for robot-based trading.
The project, based on data collection from stakeholders in the industry, relevant documents and report, is analyze the role of digital data platforms as vehicles for operating current and future energy markets
This project is tied to the PowerDig project,
https://prosjektbanken.forskningsradet.no/project/FORISS/320789?Kilde=FORISS&distribution=Ar&chart=bar&calcType=funding&Sprak=no&sortBy=date&sortOrder=desc&resultCount=30&offset=90&TemaEmne.1=Portef%C3%B8lje+Naturvitenskap+og+teknologi&source=EU&projectId=965417
See background from NTNU's Energy Transition, https://www.ntnu.edu/energytransition
However, in many cases the ISA is either undocumented, unknown, or unavailable. In addition, malware has been shown to use custom virtual machines to avoid detection. Such cases prove extremely time intensive for the reverse engineer. ISA features such as word size, instruction format, register size, and number of physical registers are a prerequisite to disassembly.
The task for this project is to develop a heuristic method to analyse binary programs and extract instruction set architectural features such as:
Knowledge of computer architecture and assembly programming is helpful but the lack of such knowledge should not discourage the applicant(s). Strong Python programming skills are desirable.
The Approximate Nearest Neighbor (ANN) algorithm is a popular algorithm used in machine learning and data mining. It is used to find the nearest neighbor of a given point in a high-dimensional space. However, as the size of the dataset grows, the performance of the ANN algorithm degrades. This is because the ANN algorithm requires a lot of memory and computation power to work efficiently.To address this issue, researchers have proposed distributed versions of the ANN algorithm that can scale beyond the 1 billion vector mark. However, these distributed versions are not efficient enough to handle large datasets. Therefore, there is a need to build a distributed version of an ANN algorithm that can offer efficient scaling beyond the 1 billion vector mark.Your thesis will explore how to build such a distributed version of an ANN algorithm that can offer efficient scaling beyond the 1 billion vector mark. You will investigate different approaches to building such an algorithm and evaluate their performance on large datasets. You will also explore how to optimize the performance of the algorithm and how to make it more efficient by using different strategies for cross machine partitioning and communication.
This project will require extensive programming, and students should have excellent skills in C++ and Python to take this on. Furthermore, experience and good capabilities in performance evaluation and understanding of detailed performance characteristics of computers will be required.
Project objectives:
Programming language: Python
Skills required: machine learning, deep learning, object detection, and localization models in images/videos.
Problem statement:
To support the end-to-end (E2E) learning of a driving system, a simulation environment should ideally:
As of today, there is no open-source simulator that truly covers all three properties:NuScenes, Waymo Open, KITTI and other pre-recorded datasets are limited to offline/open-loop evaluation.The CARLA simulator simulates a proper closed-loop environment, but its scenarios are scripted and not very complex.Finally, the nuPlan simulator supports both closed-loop evaluation and complex real-world situations, but sensor inputs are only available in the open-loop setting, not in closed-loop simulation.
Approach:
Recently, novel view synthesis methods (Neural Radiance Fields, 3D Gaussian Splatting) have become very successful.Such methods can generate high-quality RGB images of a scene from previously unseen positions and view angles. This is precisely what nuPlan is missing.
The goal of this project is to explore, how such view synthesis methods can be used to extend the nuPlan simulator by giving it the ability to synthesize camera-inputs on the fly.
Sound interesting? Don't hesitate to reach our.
Project Goal: The project aims to evaluate various LLMs (ChatGPT, Palm, LLaMA, Mistral) and compare their performance in generating a set of questions using prompt engineering techniques.
Description:
Train different LLMs on the complete contents of different courses/degree programs, for example, BS Computer Science/ MS in Informatics or a single course. Contents could be all the books related to different courses in the degree course schema. Once a model is trained, we prompt the model to create a question paper. The creation of a question paper can be based on a description provided by the instructor. For example, the structure of the paper, the number of questions, their types (short, long, MCQs)., as well as from the perspective of the level of difficulty, like memory level questions, understanding, creativity, and so on. Once the paper has been formulated, assess the LLM's ability to score students' responses to those question papers. The project may further explore LLMs' ability to provide students with descriptive feedback. The overall idea is to assess different foundational models' capability in e-assessment and assess the impact of fine-tuning those foundation models.
1. Train different LLMs on a course/study program. - material can be used from IDATG2204 course content available on Bb.
2. Use prompt engineering techniques to generate question papers.
3. Evaluate the responses obtained on the question paper and provide feedback.
Challenges: The major challenge is to train LLMs for a study program
Since its maiden release into the public domain in November 2022, ChatGPT garnered more than one million subscribers within a week. The generative AI tool ⎼ChatGPT took the world by surprise with its sophisticated capacity to carry out remarkably complex tasks. The extraordinary abilities of ChatGPT to perform complex tasks within the field of education have caused mixed feelings among educators as this advancement in AI seems to revolutionize existing educational praxis. This topic will investigate the opportunities and challenges of generative AI (through the implementation of a case study) and offer recommendations on how generative AI could be leveraged to maximize teaching and learning. The goal is to identify how these evolving generative AI tools could be used safely and constructively to improve education and support students’ learning.
The project aims to revolutionise university credit management by integrating blockchain with existing educational systems using Blackboard and Inspera APIs. EduWallet ensures secure, efficient record-keeping and easy credit transferability across institutions.
API Integration:
Smart Contracts for Enrollment: Simplifies the course enrollment and withdrawal processes using blockchain to guarantee secure, verifiable transactions.
Credit Transfer: Employs a blockchain ledger for secure storage and seamless transfer of credits, enhancing student mobility.
Real-Time Verification: Offers instant verification of academic records, accessible by employers and institutions, ensuring accuracy and preventing fraud.
Token-Based Incentives: Rewards students with tokens for academic achievements, redeemable for special privileges like exclusive workshops or priority course enrollment.
Analytics Dashboard: Provides a real-time dashboard for students to monitor credits, course statuses, and rewards eligibility.
Web app (and or Mobile app), Blockchain, API integrations, Database, Security
With the rise of functions as a service (FaaS) due to the broader availability of cloud computing workloads, we face an evergrowing body of software written in so-called high-productivity languages, such as Python or Javascript. While we have a somewhat detailed understanding of how compiled software performs on modern systems, these interpreted programming languages pose unprecedented challenges. All these languages require a runtime consisting of compiled code which interprets the actual logic and translates it on the fly into an intermediary representation, similar to the bytecode representation of C# or Java.
The system sees the runtime as regular code and thus uses the hardware structures designed to handle code. However, the actual functionality of a program written in an interpreted language is considered data from the hardware perspective. Therefore, it is handled by hardware designed to manage data. The effects of that behaviour on the microarchitectural state of a processor, the memory hierarchy, and the entire system's performance need to be better understood.
To achieve the level of understanding needed, we require better tooling to measure the behaviour of such workloads throughout the computing pipeline. We need more than the currently existing tools to track the data required to assess those effects. The first part of the project is concerned with designing and exploring the space for such a tool that can precisely pinpoint which cache lines contain actual data that the interpreted program is operating on and which cache lines have code to be interpreted by the interpreter.
While the tool helps us track the information in the front end, we must understand its effects on the decoupled front-end and back-end aspects. While the back end is most likely only indirectly affected by the challenges that interpreted languages pose (e.g. stalling introduced by reducing the effectively available L1D cache), the front end is expected to be affected very directly, starting with the inefficient use of caches and the wrong assumptions of the replacement policies, to the prefetchers that are geared towards data or instructions, to the branch predictors, that is lost because the interpreters branching behaviour mostly depends on data and thus is hard to predict.
Understanding such problems' behaviour is of utmost importance to address those challenges. With Python and NodeJS being the two single-most used frameworks for the Function as a Service paradigm and PHP still being the driver for many sizeable open-source back-end software projects, understanding and optimizing hardware towards these types of programs is vital. Understanding the execution allows us to propose new methods that not only ensure the efficient execution of such programs to meet performance requirements but also targets the energy-efficient execution of workloads on cloud computing infrastructure.
This thesis aims to explore the use of DiskANN for batch processing and searching over immutable data. The first step will be to build a good Python interface for DiskANN that can be used to experiment with the algorithm. The thesis will then investigate how to optimize the performance of DiskANN for batch processing and searching over immutable data. This will involve exploring different strategies for cross-machine partitioning and communication.The thesis will also evaluate the performance of DiskANN on large datasets and compare it with other state-of-the-art approximate nearest neighbor algorithms. The results of this evaluation will be used to identify areas where DiskANN can be improved and to suggest directions for future research.This project will require extensive programming, and students should have solid experience in C++ and Python to take this on. Furthermore, experience and good capabilities in performance evaluation and understanding of detailed performance characteristics of computers will be required.
This project is developed in collaboration with a company. The main idea of this task is to minimize the time the company spends to generate MVPs from a given idea of a product.
The project will focus on developing a solution for the reuse of services and components developed by the company, abstracting customization patterns and variability aspects. The project aims at developing a prototype tool that fully customizes and assembles existing components from a customized design-time view of the system, such as a prototype design made in, e.g., Figma. Additionally, the user should be able to import services to the project through the tool.
The scope of the project is a "proof of concept" of this tool, as well as an analysis of possible methodologies for reusing services and components in this context.
Emotion are essential in work and learning. For instance, positive emotions can lead to more creativity, and emotions can influence memory and decisionmaking. In technology enhanced learning, recent years have seen a strong interest in emotions during learning.
This task will start by reviewing the literature about how emotions and well-being have been addressed in computing education. The task can then continue with the collection of empirical material to understand better the current needs of students (with IDI as a case).
Depending on time, this understanding might then be used to inform the design of a tool for improving awareness of emotions in computing education. The tool might be a game or another form of light-weight tool.Students are welcome to discuss specific areas of interest, with respect to both topic and supporting tool.
Contact the supervisor to share your ideas and know more about this task.
Supervisors: Michail Giannakos, Giulia Cosentino
Place: LCI Lab: https://lci.idi.ntnu.no/ and in collaboration with UC Berkeley’s EDRL lab (https://edrl.berkeley.edu/)Suitable for: Two students
You can see a video of the current prototype here: https://www.youtube.com/watch?v=3tpqfkSDkE8
IntroductionThe field of multisensory technology allows the learner to interact beyond usual input devices (e.g., keyboard and mouse) with new forms of natural user interfaces such as full body interaction, gestures, vocal etc. and provide multisensory stimulations (e.g., through sounds, lights, visuals) to support users' learning. Moreover, the number of sensors provided by the multisensory technologies supports the collection of multimodal data that add sensemaking and predictive powers to previous forms of analysis. The goal of this project is to explore how Motion-Based Technology (MBL) and Gen AI capabilities (e.g., LLMs and Multimodal LLMs) may facilitate children’s meaningful STEM learning.
Thesis DescriptionIn the first step, the student needs to review the literature and familiarize with the adaptive learning models, databases and processes and MMLA understanding also the context of multisensory technologies. Then, the candidate supported by the best practices found and adapted from the literature, will define the model able to suggest the activity and interaction the learner will have to perform with a given content and interaction paradigm, based on the MMLA. Overall, the project aims to: 1) identify aspects of Motion-Based Technology (MBT) and analytics that support children’s learning, with a focus on math, 2) develop a set of practices, functionalities and technological interfaces that materialize those aspects, and 3) evaluate and refine those practices and the technological interfaces.
RequirementsThe ideal candidate will have a background in data science and modeling, and experience with using/integrating LLM technology to applications. Solid back-end programming skills (Python or C# or JavaScript) and an interest in hands-on development and experimentation is also a requirement.
Programming skills: Python or C# or JavaScript
This project explores the significant challenges and proposes innovative solutions within the field of Retrieval-Augmented Generation (RAG) techniques, particularly focusing on domain-specific applications. One of the primary challenges addressed is the development of robust embedding vectors that accurately capture the nuanced language and terminologies unique to specific domains. This involves enhancing the algorithms that process and understand textual content, thereby improving the model's ability to integrate and utilize domain-specific knowledge effectively.
Another critical aspect of this research is improving the precision of citations generated by these models. As retrieval-augmented generation techniques pull information from various sources to produce content, ensuring the accuracy and relevance of citations is paramount. This project aims to refine the retrieval mechanisms and citation algorithms to increase the reliability and verifiability of the generated texts.
Through a combination of empirical research, algorithmic enhancements, and practical testing, this thesis aims to advance the field of retrieval-augmented generation by addressing these key challenges.
Objective: To develop and evaluate AI-based analytics models for tracking and enhancing learner engagement and performance in simulations, focusing on STEM.
Description: This research aims to develop AI-based analytics models to enhance the effectiveness of laboratory simulations in online learning. The study will focus on tracking learner interactions within simulations developed using Articulate Storyline 360, generating actionable insights into student engagement, skill acquisition, and performance trends. These insights will inform real-time interventions and instructional design strategies to improve student outcomes. A case study on STEM education.
Cities are evolving rapidly due to technological advances such as digital ecosystems, sensor technologies and vast amounts of data. At the same time, the need for evolving in a sustainable manner and ensuring innovative and sustainable solutions are of utmost importance. Cities stand to gain from learning from other cities. Enterprise Architecture has been considered as a means of capturing an ICT ecosystem in a city so that it can be replicated in other cities. This project will also focus on if and how ideas from enterprise architecture could be applied to support the transfer of knowledge and learning across cities. The tasks will include the following:- Literature review- Enhance enterprise architecture model to support learning and knowledge transfer- Prototype model - Validation of the model.The courses in information modelling and Enterprise Architecture and Innovation (TDT4252) and the specialisation module TD20 (Smart and Sustainable cities and Enterprise Architecture) will be relevant for this project.
Smart and personalized systems such as recommender systems (or artificial intelligence in general) keep influencing our daily lives in an increasing rate. In the recent years, researchers became more aware of the ethical challenges in developing such systems in an ethical way such that these systems would treat everyone equal and fair, without any bias or discrimination. Even though these are the topics social scientists have been working on for a long time, defining these concepts as mathematical models, implementing them within AI systems and evaluating the success of these approaches is not an easy task.
This project focuses on the ethical aspects such as privacy, fairness and bias in recommender systems from the technical point of view. The direction and details of the project can be clarified upon a discussion with the students.
A few references on the topic:
https://blog.fiddler.ai/2021/03/ai-explained-understanding-bias-and-fairness-in-ai-systems/
https://www.ibm.com/blogs/research/2021/04/ibm-reduce-bias-in-healthcare-ai/
On March 4, 2024, Finansavisen published a piece on Anna Asset Management (AAM) after the hedge fund Anna Fund posted a 63,5 % return in February. AAM, founded by former students at the Department of Industrial Economics and Technology Management, NTNU, trades based on an algorithm designed to exploit momentum in the Bitcoin market.
According to the efficient-market hypothesis (EMH) it should be impossible to consistently beat the market with such strategies. However, cryptocurrency markets may not be efficient to the degree that predicting price movements is hopeless.
There is an evolving literature on the efficiency of cryptocurrency markets (see e.g. Yang, Jeong et.al (2023)*). This project aims to evaluate the degree of market efficiency of cryptocurrency markets over time using novel artificial intelligence models, such as Temporal Fusion Transformers, that may detect patterns that traditional econometric approaches struggle to identify.
The project will be co-supervised by Einar Belsom.
Objective: To create gamified simulations with adaptive AI mechanics in Articulate Storyline 360 and evaluate their effectiveness in enhancing motivation and promoting deeper learning.
Description: This topic investigates the use of gamification in simulation-based learning environments developed with Articulate Storyline 360. The study will incorporate AI elements such as performance tracking and adaptive game mechanics to personalize learning experiences. By examining learner motivation, engagement, and retention rates, the research will assess how gamified simulations can address challenges in online education and promote deeper learning.
In AI and ML, several methods rely on stochasticity or randomization: mutation and crossover in evolutionary algorithms; dropout and stochastic gradient descent in deep learning; stochasticity in stochastic local search; and randomization in systematic search. Evolutionary algorithms (EAs), which we study here, are competitive in solving many computationally hard problems while modeling important biological phenomena. Further, EAs are interesting in that they can studied formally, for example by means of Markov chains.
Project Focus
The goal of this project and thesis is to study the theory and application of EAs. The following research questions are examples - specifics depend on finding a topic of broad and common interest. First, there are opportunities to improve EAs, with an eye to specific applications. Can this done by combining different heuristics and adaptive methods, perhaps including concepts from algorithms for stochastic local search?
Second, the computational cost of ML can currently be high, and the use of EAs for ML is no exception. Maybe Markov chain-based analysis and/or high-performance computing can be the basis for EA-based ML methods that better handle massive and complex datasets while reducing computational cost dramatically?
Third, the theory and formal methods communities have studied parametric Markov chains, but interactions with the AI and ML communities have been limited. Perhaps this will change, now that trustworthy AI is coming to the forefront?
Notes
In this project, the joint interests of (potential) sponsor, advisor and student(s) come together. Typically, the project will be based on the problem described by the sponsor AND previous research by Prof. Ole Jakob Mengshoel AND interests of students. If only one or two of these are present, there is no basis for a project.
Metabolisk syndrom er et enormt helseproblem som påvirker stadig fler i den vestlige verden. De siste tiårene har gjort oss mer stillesittende og gitt oss enklere tilgang på ultraprosessert mat samtidig som jobbene våre har blitt mindre aktive. Resultatet er at stadig fler av oss havner under kategorien metabolsk syndrom. Dette øker risikoen for å utvikle hjerte og kar sykdrommer samt diabetes type 2.
Dette arbeidet fortsetter tidlig arbeid hvor en Genetiske Algoritme produsere kontinuerlige og dynamiske treningsprogram som utvikler seg i takt med pasientens helsetatus og hjelper dem å forbli friske over lang tid.
If you have a clear idea of an exciting topic you want to work on in HPC (high-performance computing), including Parallel Computing, Quantum Computing simulations and Cloud Technologies, contact me (Anne Elster elster@ntnu.no) to see if we are a fit and how feasible you project is as a fall project and/or Master thesis.
Topics include investgating new parallel algorithms, GPU features, including testing NVIDIA B100s,and other new HPC systems.
I am aslo interested in developing edutainment for teaching parallel programming and parallel computing!
AI-technology has revolutionized many business processes. In banking, organizations increasingly rely on complicated models such as deep neural nets. These models suggest which products or services to advertise to customers.
While the models might give optimized suggestions, employees need to understand their reasoning. They ask “why should we recommend this product to the customer?”
Explainable AI constitutes a growing research area. The candidate will explore different methods to examine models' output and add explanations.
Sparebank 1 SMN will provide a large data set with up to 200,000 records and 2,000 variables. The candidate will have access to a laptop with which they can access the data and run experiments on a virtual machine.
Stian Arntsen (Sparebank 1 SMN) and Inga Strümke (NTNU) will support the master project.
The project will mainly be based on simulations in mumax3 as well as our open-source flatspin simulator for large-scale ASI systems.
The goal of this project is to tailor the switching characteristics of nanomagnets for computation. This is an interdisciplinary project at the intersection between computing and physics.
ASIs are currently being investigated as a promising substrate for material computation. To this end, the COMET group recently published a novel method to interact with ASI systems known as “astroid clocking”. Astroid clocking is a powerful way to supply input to and control the dynamics of an ASI.
Astroid clocking exploits the shape of the nanomagnet switching astroids’ within the ASI. As such, it is of great interest to know what astroid shapes are possible and how they can be achieved. There is little published knowledge and data on what astroid shapes are possible and how to achieve them, aside from a few canonical examples.
Elongated nanomagnets behave as binary spins, i.e., with two stable magnetization directions. A nice feature for computing, we can treat the state of these magnets as binary, “1” or “0”. A magnet will flip between these two states when a strong enough magnetic field is applied. The required field strength depends on the angle at which the field is applied, and this relationship is described by the magnet’s switching astroid.
The switching astroid of a nanomagnet is governed by the magnet’s shape and size. This aim of this project is to investigate what astroid shapes are obtainable. This could be exploring the possibilities with an evolutionary novelty search sort, or otherwise. Alternatively, a method of searching for a suitable magnet shape given a target astroid shape could be developed. In either case, the project involves running micromagnetic simulations of magnet shapes to determine their respective astroids.
AI governance is a notion that is often attributed to a range of different practices and processes. From establishing a process of developing AI applications, ensuring that quality outcomes are achieved, and to decising the role and responsibilities of stakeholders. AI governance now plays an important part related to the business value that AI can deliver, and to ensuring that projects comply with ethical and regulatory frameworks. This project will seek to understand how organizations develop AI governance practices, what aspects they take intio accoutn when doing so, how they deploy them, and what the outcomes of them are at the business and project performance levels.
The focus of this thesis is to develop a system to help the novices in programming while debugging. One of the ways to provide help is to learn from the expert about how to look at the program while finding the bugs in the code. This way of providing help is called Expert’s Movement Mapping Examples. Most of the efforts in this direction include the use of expert’s gaze in the problem space. In this thesis the student(s) will exploit the use of dialogue as well as the gaze of the expert.
Thesis DescriptionIn a first step, the student(s) will design and implement the gaze-enabled feedback tool. Afterwards, they will conduct a small user study in order to test the usability of the system with a small number of students. Once the usability of the system is established (with the last changes in the system), the student(s) will conduct a larger user study to evaluate the effectiveness of the system. Finally, the candidate(s) will analyse the collected data and write up his/her thesis.
RequirementsThe ideal candidate will have a background in system design. Solid programming skills and an interest in hands-on development and experimentation is also a requirement.Programming skills: Python/Java.
Detecting facial landmarks in 2D facial images is a prerequisite for expression transfer, among other applications. This project will survey existing approaches for 2D facial landmark detection and implement a selected technique. Applications will be considered, such as expression transfer.
https://paperswithcode.com/task/facial-landmark-detection
Knowledge: Python, C/C++
Dr Antonios Danelakis, Researcher, IDI, NTNU antonios.danelakis@ntnu.no
The impact of news articles on the society can not be underestimated and as the number of online news are increasing, distinguishing the fake news from real news is becoming a challenge for people. This project focuses on analyzing and/or tracking news articles from different news sources or social media channels, in order to find an efficient way of detecting fake news.
In this project different methods (such as machine learning, natural language processing, semantic web etc.) or approaches can be used towards the detection of fake news / disinformation.
The details of the project can be clarified upon a discussion with the students.
In this project, the aim is to address the challenges of analyzing medical images while maintaining patient privacy. In this approach, instead of centralizing all medical data in a single location, the project employs federated learning, a decentralized machine learning technique. In a federated learning setup, multiple medical institutions or entities collaborate to train a shared machine learning model without sharing raw data. Each institution retains control over its data, and only model updates are exchanged and aggregated among the participants. This approach ensures that sensitive patient information remains localized, minimizing the risk of data breaches and privacy violations. The project's focus on medical image analysis indicates that it's aimed at tasks like diagnosing diseases, detecting anomalies, or segmenting regions of interest within medical images. By using a collaborative federated learning approach, the project combines the benefits of data sharing for model improvement while upholding strict privacy standards required in the medical field.
Large language models form the basis of almost all currently topical AI research, making it vital to identify and rectify different types of demographic biases in those models (be that bias based on gender or sexual identity, or on cultural, ethical or social background, etc.). This has triggered intense research on fair representation in language models, aiming both at building and using unbiased training and evaluation datasets, and at changing the actual learning algorithms themselves. There is still a lot of room for improvements though, both in identifying and quantifying bias, in developing dibiasing methods, and in defining bias as such.
Wildfires pose a significant threat to ecosystems, communities, and infrastructure. This project focuses on the development and enhancement of wildfire detection and early warning systems using AI and computer vision. By leveraging a combination of remote sensing data, sensor networks, machine learning, and artificial intelligence, the study aims to create a robust and proactive solution for early wildfire detection. The research delves into the challenges of accurately identifying wildfire signatures from various data sources, optimizing real-time data processing, and designing efficient alert mechanisms to enable timely response and mitigation efforts. The outcomes of this research contribute to minimizing the impact of wildfires on lives and the environment through advanced fire detection strategies.
Multiple movements like opening or closing the hand, grasping, or showing the palm can be decoded from the EEG signals recorded while attempting to do those movements. The decodified movements can serve for multiple purposes. For example, in neurorehabilitation they can be used to provide feedback to a patient that is performing therapy to recover hand movements after stroke, and in brain-computer-interfaces to generate outputs that control an external device such as home appliances, computer games, toys.
The objective of this project consists of decoding movement intentions by combining low-density EEG and source reconstruction (estimation of the activity inside the brain from the electrodes on the scalp). The project involves recording EEG signals for multiple participants, analyze and build offline/online classifiers using state-of-the-art machine/deep learning algorithms and developing software games such as the one in this link: Video semi final (youtube.com).
Preliminary results are now published here: https://link.springer.com/article/10.1186/s40708-024-00224-z
This project will provide a foundation to develop wearable solution based on few electrodes that can be later use in neurorehabilitation therapies. The project is done in collaboration with Marta Molinas at the cybernetics department
Visuospatial neglect are commonly experienced neuropsychologicalconditions affecting the contralesional side in post-stroke patients, leavingpatients with egocentric or allocentric perceptual problems. Diagnostic tools forvisual neglect include the apples test, balloons test, and bells cancellation test.All administered on paper. While psychometrically sound, these tests areadministered in an overtly clinical setting, lacking depth as a test parameter,only allowing for crude temporal data collection and gaze observation, andalso being limited in spatial scope to the size of the paper. By having a limitedset of test parameters, the status quo represents a barrier to advancing ourunderstanding of the mechanisms behind visouspatial neglect and its effect ineveryday settings of the patients. To mitigate these limitations, a VRenvironment is being developed for assessment of neglect by utilizing low-cost,off-the-shelf, VR headsets with integrated eye trackers, along with a custom-developed and highly flexible virtual environment, where the test parameterscan be altered based on the needs of the clinician.This master project aims to bring flexible data visualization into theaforementioned VR environment. For starters, the students will look at usingPython to generate graphs/plots of gaze data which has previously beencollected from the VR environment, and then to display these plots directly inthe VR environment itself (i.e. the gaze plot of the person during the test).Next, one can classify the fine-grained gaze data into more meaningful unit,such as fixation or saccade, and come up with more relevant staticalinformation to the clinician. Eventually the information should be visualized inan easy to comprehend manner to the clinician.This work collaborates with Department of Acquired Brain Injury, St. Olav'sHospital.
Supervisor: Alexander Holt
Co-supervisors: Tor Ivar Hansen, Xiaomeng Su
The project aims to advance the understanding and interpretation of three-dimensional (3D) point cloud data through segmentation techniques. Point cloud data represents objects or scenes in 3D space using individual points, often collected from technologies like LiDAR sensors or depth cameras. Semantic segmentation in this context refers to the task of labeling each point in a point cloud with a specific class, such as "car," "tree," or "building." This enables the identification and categorization of objects within the 3D environment. Panoptic segmentation, on the other hand, extends this concept by not only labeling object instances but also recognizing 'things' (e.g., individual objects) and 'stuff' (e.g., surfaces or background). The project aims to develop a comprehensive approach that bridges the gap between semantic and panoptic segmentation for point cloud data. This involves creating models and algorithms capable of not only recognizing object categories within the point cloud but also differentiating between individual instances of those objects and understanding the overall context in which they exist. The potential applications of this project are significant. It could contribute to advancements in autonomous driving by enhancing the ability of self-driving vehicles to perceive and react to their surroundings accurately. Additionally, it could be useful in various fields where 3D data analysis is critical, such as robotics, augmented reality, urban planning, and environmental monitoring.
The immune system is arguably one of nature's most highly adaptive, distributed and self-organising systems. It has the property of being able to recognise anomalies --- something that deviates from the common rule. In 2018-2019 two masters students proposed a novel hybrid classification algorithm MAIM, combining such features of the Immune Systems with an Island Model Genetic Algorithm (IGA). The preliminary results achieved are promising and resultet in an international publication. The goal of this project is to build on this foundational work and investigate the many avenues available to further extend/refine this novel algorithm.
Contact Pauline haddow, pauline@ntnu.no for further information
As a result of increasing digitalization, IT tools are used in multiple contexts by users with multiple backgrounds, and are intertwined in complex ways with everyday practices. The complexity of the context of use arises a number of ethical issues for software developers and users, including e.g., algorithmic biases, privacy of personal data, addictive design, …
This specialization/master project aims at the development and evaluation of a mobile game to be used in events, workshops, and courses, with the aim to facilitate reflection, and discussion on ethical issues related to IT tools.
Previous work has been done in the group about teaching about sustainability to computer science students and provides a good starting point, still giving freedom to shape your work.
Supervisors: Letizia Jaccheri and Claudia Maria Cutrupi
This project/master thesis will build on the existing bulk of knowledge about gender and diversity in software development (TDT10) to provide increased knowledge and solutions about Inclusion and Diversity, especially gender diversity, in Computer Science over Europe. Specifically, in this project/master thesis, the student(s) will propose one or more goals to investigate as discussed below:
The student(s) will - analyze gender balance and diversity in Computer Science across Europe,
- examine and explore Computer Science role models with a focus on diversity,
The project is conteztualized in EUGAIN (European Network for Gender Balance in Informatics), which has more than 150 members. Materials to analyze are already collected and available, including videos produced by role models in CS and video reports about the situation in European countries. For an example, see the project’s YouTube and TikTok.
The supervisors will provide the student(s) with Initial Literature and help the student(s) to access to Stakeholders and initial data for the Empirical Investigation
Resources
https://sbs.idi.ntnu.no/
co-supervisors Anna Szlavi and Letizia Jaccheri
This project/master thesis will build on existing bulk of knowledge about gender and diversity in software development (TDT10) to provide increased knowledge and solutions about Inclusion and Diversity, especially gender diversity, in Computer Science over Europe. Specifically in this project/master thesis, the student(s) will propose one or more goals to investigate as discussed below.
The student(s) will - analyze data in connection with the gender aspects of Computer Science education,
- contribute to developing a Leadership and Inspiration Academy at NTNU to facilitate gender-inclusive education,
- design, implement, and evaluate new tech solutions which contribute to solve the problem of inclusion in Computer Science education (for example, developing an app or a website for university students)
Students will work within the frames of the project Women STEM UP (Home – Women Stem Up (women-stem-up.eu). Materials to analyze are readily available, including surveys collected with NTNU students and teachers regarding the state of CS education and gender.
Some possible practical outcomes are: developing a mobile app or a website for mentoring, gender awareness, gender training in CS, etc.
The supervisors will provide the student(s) with Initial Literature and help the student(s) to access to Stakeholders and initial data for the Empirical Investigation Resources
SINTEF Digital, Cognite, Kongsberg, NorwAI: In hybrid AI the aim is to combine machine learning and physical knowledge. This is particularly relevant for industrial applications of AI and machine learning on physical systems. Through the SFI NorwAI we work on hybrid AI together with partners in the energy sector. We have two use cases focused on predictive maintenance of wind turbines and on virtual flow meters in oil and gas.
Simulation based inference is a likelihoo-free Bayesian inference method where one uses machine learning to approximate the likelihood function based on simulation data. This allows for Bayesian inference when no analytic model exists. The method naturally performs best in cases of stochastic models. The project can either have an applied or a theoretical focus. For the applied focus, the goal is to combine existing commercial black box simulation software and real data from wind turbines and compressors to speed up model identification (which input parameters are required for the simulation to match the data) and quantify the model uncertainty. For the more theoretical focus, the goal is to study different machine learning models for probability distributions and how they can be constrained to meet certain physical constrains.
Thesis Description:
The overall learning outcomes will be
The project can be adapted to students from several departments such as Computer Science, Mathematical Sciences or Engineering Cybernetics.
Data Description: The project is based on a mix of public benchmark data and proprietary simulations from industrial partners (DNV for wind turbines and Kongsberg for compressors). This will require approval from the partners before publications.
Supervisors: Alexander Stasik (SINTEF), alexander.stasik@sintef.no, with input from Andris Piebalgs (Cognite), Even Pettersen (Kongsberg), Jon Espen Ingvaldsen (NTNU)
With the advent of neural networks, signed distance fields (SDFs) and ray fields are gaining traction as object representations. But since rendering hardware is based on surface representations, it is often necessary to extract such a representation from the above fields. The classical (but costly) marching cubes approach is often used for SDFs. This project will investigate more clever ways of creating a surface from an SDF and possibly its gradient, as well as from Ray intersection fields, such as the MARF (Medial Atom ray Field). The investigation will start with genus 0 closed objects and SDFs.
The MARF was developed at NTNU by Peder B. Sundt.
Peder Bergebakken Sundt, Researcher, IDI, NTNU peder.b.sundt@ntnu.no
Theoharis Theoharis, IDI, NTNU theotheo@ntnu.no
Generating synthetic data. Banking data can contain sensitive information that limits the possibility of sharing the data openly. The data is typically restrained to being analyzed on our internal hardware. This limits the ability to use large compute clusters and cloud services. It also sets a high barrier of entry for using the data in research than what an open dataset would. If synthetic data could be created based on real data, which retains the inference and statistical relationships without containing personal data, this could make it both easier to share data and to use compute resources outside of our servers.
Thesis Description: Methods for generating synthetic data include generating data from statistical distributions, using agent based modelling, GANs and LLMs. Text data that contains personally identifiable information is hard to synthesize. Transactional data, such as card transactions can be easier to remove any personal information from, but it will probably be hard to maintain statistical relationships, especially in tandem with other customer data. One approach could be to try to synthesize an already aggregated dataset, that would contain for example the sum of transactions for a given category such as restaurants. A younger person might have a very different history of transactions. In that case, a GAN should be able to create datasets that retain the relationship between age and restaurant expenses. Another approach could be to synthesize different raw transactional data, that is connected to customer data and product data in a relational form. In that case, it would be more challenging to generate data where these statistical relationships are retained, but this would also open up a lot more use cases than synthesizing a single table or flat file.
We expect the student to synthesize banking data in such a way that the data will retain statistical information while not containing any real data. The synthesized dataset should give the same statistical inferences as the original data.
Data Description: The data would either be an already aggregated flat file, for example one of our analytical base tables for next best offer models, or a subset of relational data, such as transaction data, product data and customer data. The latter would be more challenging but also open up a lot more use cases. Students working on our data will be working on a virtual machine through a company laptop. All the data processing will have to be done on the virtual machine. The data will be de-identified.
Contact: Stian Arntsen, Stian.arntsen@smn.no
SpareBank1 SMN is the leading financial institution in middle Norway. SpareBank 1 SMN is an independent regional savings bank with a local footing and one of six owners of SpareBank1-alliansen. SpareBank 1 SMN offers competitive products in the fields of financing, savings and investment, insurance and payment services along with estate agency, leasing, accounting and capital market services. The thesis will be carried out in the advanced analytics department in the technology and development division. Relevant business resources will be included if necessary.
Supervisor (NTNU): Jon Espen Ingvaldsen, jon.espen.ingvaldsen@ntnu.no
Generativ kunstig intelligens gir oss muligheten til å være mer produktive i programmering og prosjektarbeid ved at vi kan få automatisk generert mye kode eller forslag til kode mens vi skrive. I kontekst av læring gir teknologien både utfordringer og muligheter og i denne oppgaven kan du utforske eller prøve ut forskjellige løsninger for bruk av generativ AI relatert til programmering og programvareutvikling. Tema som kan være relevante er:
The software codes must be secure for business- and mission-critical systems (e.g., IT systems in banks) because many people will use them for a long time. Although existing GAI models can generate code, the generated code might be insecure due to vulnerabilities in code training the model. This project aims to identify the challenges of applying GAI to achieve software security and propose approaches to incorporate security requirements, principles, and optimizations as more explicit knowledge into GAI models.
The tasks include:
+ Literature review on the challenges of GAI on software security.
+ Propose approaches to fine-tune GAI models to make it possible for them to generate more secure code
+ Evaluate the proposed approaches with open-source projects
Diffusion models (see e.g. these lecture notes) have gained quite some interest lately, with impressive results in particular in the image domain (now also doing high quality video from text), but also used for a lot of other things including reinforcement learning and generation of both time series data and graph data. Diffusion models have also inspired more fundamental research into alternative formulations of generative AI, including Bayesian flow networks.
In this project you will work with models for generative AI. The content of the project can be adapted to your interests: A purely theoretical task, a comparative study (e.g., a shootout between different versions of diffusion models, comparing to GANs or VAEs, etc) or a practical one related to a dataset of your interest are all possible.
Text Mining constitute a field in Artificial Intelligence that parses large collections of documents to detect valuable patterns. Medical documents comprise a plethora of useful information. The information can help to develop better medical interventions or prevent diseases. The master project aims to develop a system that takes in medical documents and extract valuable information. For instance, the system could find relations and produce a knowledge graph.
The candidate should have attended courses about Artificial Intelligence, Machine Learning, and Natural Language Processing. Ideally, the candidate should have vested interest in the topic. The candidate should have sufficient programming skills to process the documents, implement a set of algorithms, and develop a demonstrator that showcases the system's capabilities.
A good relationship with their customers is essential for financial institutions. This involves reacting timely and adequately to customer messages. AI technology has the potential to improve customer services and become a competitive advantage for banks.
Many financial institutions rely on established machine learning models. Recently, Large Language Models (LLMs) have exceeded previous performance in many benchmarks. The candidate's task is to compare a set of LLMs to simpler models to see whether they can categorize messages more reliably.
Sparebank 1 SMN will provide a data set with more than 400,000 emails and 160,000 chats mostly in Norwegian that can be accessed with a laptop provided by them.
Stian Arntsen (Sparebank 1 SMN) will support the master project.
Naturally, organizations collect large document archives. These feature a variety of documents, some of which lack important metadata. Examples include document type, and references to products, services, or customers.
The candidate's task is to enrich the document archive by applying generative AI methods to infer missing metadata. The project requires setting up an NLP pipeline to automatically process documents into a numeric representation, and apply a set of machine learning methods to obtain estimates for missing metadata with high confidence.
The candidate will get access to a document archive from DNB and can assemble a reference data set from public sources. The project will be supported by Jan Thomas Lerstein, DNB.
The candidate will have the opportunity to engage in a summer internship at DNB in Oslo to prepare for the project. They can work on the thesis in Oslo.
I programmerings og prosjektfag hvor studenter gjør utviklingsarbeid benyttes git for samarbeid om kode og dokumentasjon. I en utdanningskontekst ønsker fagstaben ofte overordnet innsyn i prosjektene og dette kan løses med en form for dashboard hvor en faglærer på en ryddig og informativ måte kan få oversikt over og sammenligne status og informasjon om pågående prosjekter. Noe data kan hentes fra git via API, mens andre data kan genereres ved statisk eller dynamisk analyse. Oppgaven bygger videre på andre prosjekter utført tidligere
Project aim: To develop an ad campaign generator utility using LLMs and evaluate its performance
Description: Graphic designer skills are highly important in any organization. These skilled professionals design graphic content like advertisements, brochures, pamphlets, etc. The issue is a scarcity of these skills; more importantly, they are also, more often than not, overburdened with a high number of requests from different sections of any organization for graphics material development. This thesis work is related to exploiting the potential of pre-trained LLMs and subsequently fine-tuning them for any particular organization for the automatic generation of graphic content. The idea is to input a description to the model and prompt it for graphics content generation, customized to the organization's needs and aligned with the ad campaign.
Skills required: machine learning, deep learning, knowledge of GANs and Generative AI, and LLMs.
Developing countries have limited resources for healthcare delivery hence need to make the most of resources available.
This project/ thesis is linked to the Hisp/ DHIS2 (www.dhis2.org/) initiative.
Based on open source software, Hisp aims at increasing the efficency and quality of health services by enhancing the necessary reporting of health status. Mobile technologies are crucial as, even when roads and electricity is patchy, there are mobile phones.
The approach of Hisp is pragmatic: rather than elaborate, complex 'perfect' solutions, Hisp provides simple and robust ones that have a realistic chance of uptake.
Hisp, across Africa and Asia, is implemented in about 80 countries in varying degree of completion. It is one of the world's largest systems serving patients in the Global South, measured by size of the caption population.
The project/ thesis involves empirical fieldwork in Africa or Asia on selected services of the Hisp portfolio. The purpose of the work is to identify requirements and subsequently help implement these as part of the evolving portfolio of Hisp software.
The Hisp project is managed by Univ of Oslo. This project/ thesis will be in collaboration with the UiO team.
Previous dissertations (master, Phd) and reports from the UiO archive are found here: https://www.duo.uio.no/discover (search using "DHIS2" as keyword)
A digital twin is defined as a virtual representation of a physical asset, or a process enabled through data and simulators for real-time prediction, optimization, monitoring, control, and informed decision-making. This project collaborates with prof. Adil Rasheed from Institute of cybernetics. Example of digital twins relevant to this project are: an autonomous aquarium or greenhouse, an experiment of soil movement and an experiment of overload prevention in electric cables. The master thesis will focus on the development of an VR environment for visualizing of and interacting with the digital twin.
After a brain injury, patients often undergo a long and intensive rehabilitation process to regain functionality. To improve the rehabilitation process, increase its accessibility, and alleviate the high demand for clinical supervision, exergaming for rehabilitation is becoming more and more popular. Gamifying rehabilitation exercises allow for motivating training, independent from external supervision. Recently, immersive technologies such as extended reality (XR) and virtual reality (VR) are gaining traction.
However, most such VR-systems do not provide the patients with relevant and precise feedback, even though such feedback is crucial to motivate and guide the patients through rehabilitation. Tracking the patient's movements during training can give valuable information to provide such feedback. Current approaches often use specialized sensors (Kinect, accelerometers, marker-based motion tracking) and are therefore limited in their area of application and usability. With advances in Machine Learning for Human Pose Estimation (HPE), movement tracking has become a viable alternative for motion tracking, and its usability for exergaming should be further explored.
To explore the viability of HPE for exergaming, the master thesis consists of the following points: 1. Development of a prototype VR-exergame (Unity), based on a design framework for VR-exergames in a rehabilitation setting. 2. Integration of a multi-ocular HPE model for game control and feedback.
3. Brief user testing to evaluate the usability of the system. It is therefore also important that the game is easy in its setup and calibration for the targeted user group
The nature of decision-making is changing drastically, both in personal lives and in the business sphere. An increasing amount of decisions are now based on insight that is generated through analytics. Despite this, often individuals are faced with cognitive-overload, conflicted views, or biases that result in non-adoption of insight. This project will be done in collaboration with the Big Data Observatory (https://www.observatory.no) and involve designing a study protocol and collecting and analyzing neurophysiological data (eye-tracking and electroencephalography) from study participants. This will be done with the help of an expert in such tools.
Education is a unique area for the application of artificial intelligence (AI). In this topic, the augmentation perspective and the concept of hybrid intelligence will be used to guide this work. The candidates will engage with the design (co-design or participatory design) of learning services (e.g., interfaces or other artefacts) to showcase the challenges and opportunities of hybrid human‐AI learning technologies. The six levels of automation model will be used to identify the roles of the various AI users (e.g., learners, teachers). The transition of control between teacher or students and technology needs to be articulated at different levels and related to the augmentation perspective.
This project will explore how ICT could contribute to sustainable built environments that support better health and well-being of their occupants. The work will be conducted within the SWELL project: https://www.ntnu.edu/sustainability/swell.
The tasks will include:- A literature review of how ICT could contribute to health and well-being in sustainable built environments.- A literature review of relevant interactive and ubiquitous digital technologies.- Design and prototype of a solutions to engage users of buildings, or other physical spaces.- Evaluation of the prototype.
Properly identifying hatespeech is a pressing issue for social media sites as well as for smaller companies, clubs, and organisations that allow for user-generated content. Many such sites currently use slow, manual moderation, which mean that abusive posts will be left online for too long without appropriate action being taken or that content will be published with delay (which might be unacceptable to the users, e.g., in online chat rooms).
The thesis project would look into previous efforts to identify hate speech and cyber bullying, as well as available flame-annotated datasets from chat rooms, online games, Wikipedia and Twitter, and investigate various machine learning methods to identify such language.
Proposal for master thesis at Norwegian Open AI Lab, NTNU, supported by Norwegian National Advisory Unit for Prehospital Emergency Medicine (NAKOS), and Department of Emergency Medial Communication Centre (EMCC), Division of Prehospital Services, Oslo University Hospital.
In emergency medicine, time is of the essence. Acute conditions like stroke, myocardial infarction, major trauma, and cardiac arrest, result in loss of vital functions and life within minutes. EMCC is the key to rapid first aid when the public calls them for help by providing first aid instructions, but definitive treatment requires arrival of qualified personnel and often, rapid transport to hospital. However, most cases handled by the EMCC, are of less acuity, and to aid call takers at the EMCC to decide and prioritize, they use a paper-based triage-tool. Both acute and less acute cases may need ambulance transport, and usage of available resources are high and growing. When an ambulance transports a patient, it is not available for other missions, so usage of ambulance resources must balance the current load of missions and availability for the next (unknown) acute case. The current solution involves specially trained and experienced dispatchers (Resource Coordinators – RC) in the EMCC that manually assigns each mission to an ambulance, maintaining the delicate balance of workload and contingency within geographic regions.
EMCC in Oslo is the largest in Norway and answers calls to the medical emergency phone number 113 from a population of 1.6 million (Oslo, Akershus, and Østfold). The available resources includes 29 (night) to 45 (day) ambulances based at 15 ambulance stations. Each year EMCC handles 500 000 telephone calls, and the Ambulance department executes more than 150 000 missions each year.
Data Set
The existing data set is an extract from the clinical data system used in EMCC to record each mission – (Akuttmedisinsk informasjonssystem, AMIS). Each mission has a position that has been mapped to a standardized 1000 by 1000 m grid from SSB (Statistics Norway). We will anonymize the dataset, but the SSB-grid ties each mission to more sociodemographic information, as well as historical information such as traffic, weather and climate, public events, and moveable public holidays.
Each observation is per grid per hour and includes; number of events, response intervals for both acute and non-acute missions, and idle-time.
Example information:
Use of Data – Dynamic Resource Placement and Prediction
The goal is improved ambulance response intervals by means of distributed and dynamic placement of resources, and specifically:
The main focus of this project is on the optimization part (Point 1 above). In a "sibling project," the main focus will be on the prediction of future events.
In this project, the joint interests of sponsor, advisor and student(s) come together. Typically, the project will be based on the problem described by the sponsor AND previous research by Prof. Ole Jakob Mengshoel AND interests of students. If only one or two of these are present, there is no basis for a project.
Proposal for master thesis at Norwegian Open AI Lab, NTNU, from Norwegian National Advisory Unit for Prehospital Emergency Medicine (NAKOS), and Department of Emergency Medial Communication Centre (EMCC), Division of Prehospital Services, Oslo University Hospital.
The main focus of this project is on the prediction part (Point 2 above). In a "sibling project," the main focus will be on the optimal placement of available ambulance resources.
NB: for many of the industri proposals listed below a research assistant position can be offered in additon, in some cases even a summer job.
As some if you have reported issues with the direct visual intelligence related links provided below I also include the link to all the projects pitched thought the AI lab here.
MIA (Monitoring-Insight-Action): Your data. Your digital twin. Your health (video)
MIA (Monitoring-Insight-Action): Your data. Your digital twin. Your health (text)
Rosenborg (RBK) - Analyseavd.:
Sport Analytics: Fotball (RBK) - AI and Computer Vision for video analysis (pres)
Sport Analytics: Fotball (RBK) - AI and Computer Vision for video analysis (text)
Jotun:
A Cutting Edge Tool for Analyzing Fouling on Test Panels (video)
A Cutting Edge Tool for Analyzing Fouling on Test Panels (pres)
A Cutting Edge Tool for Analyzing Fouling on Test Panels (text)
Hydro:
Use of AI for Grain Size Characterization of Aluminium Alloys (video)
Use of AI for Grain Size Characterization of Aluminium Alloys (pres)
Use of AI for Grain Size Characterization of Aluminium Alloys (text)
St.Olavs / NTNU:
AI-assisted labeling of coronary arteries from CT-data: A next step from invasive to non-invasive diagnosis of coronary artery disease (video)
AI-assisted labeling of coronary arteries from CT-data: A next step from invasive to non-invasive diagnosis of coronary artery disease (pres)
AI-assisted labeling of coronary arteries from CT-data: A next step from invasive to non-invasive diagnosis of coronary artery disease (text)
SINTEF / SVV:
Autonomous Driving (AD) under Nordic conditions (video)
Navigation of automated vehicles in nordic conditions 1: Develop AI / CV methos for generation og HD-maps / Digital Road Twin (text)
Navigation of automated vehicles in Nordic conditions 2: Identify, test and evaluate AI-algorithms for navigation of AVs with LiDAR and/or video data (text)
Aerial laser classification (video)
Aerial Lidar segmentation using deep learning (text)
Aerial Batymetric Lidar, classification using full waveform and deep learning (text)
Proxpect Drones:
DEIDAC: Drone External Inspection Detection using Algorithms and Computer Vision (video)
DEIDAC: Drone External Inspection Detection using Algorithms and Computer Vision (pres)
DEIDAC: Drone External Inspection Detection using Algorithms and Computer Vision (text)
AI, AI, AI, what a gentle lobster:Huge potensial for Norwegian food industri (lobster forming) and avery interesting CV/AI project, involving detection and tracking (objects and key points), as well as the final classification as a function of time (green=gentle, yellow=undecided, red=not gentle). We have large amounts of data, incuding high quality labelded data.
Video
Pres
Text
Emerging technologies such as virtual/augmented/extended reality (VR/AR/XR) and generative AI such as ChatGPT, Midjourney and Sora are already revolutionizing how we live and work. XR has already demonstrated significant potential in transforming educational practices by providing learners with realistic and highly engaging learning experiences. Generative AI is a powerful tool that can be used to quickly and efficiently create a wide range of educational content, including human-like text, videos, images, and even 3D models and software code. The goal of this master project to investigate if the combination of these technologies can contribute to creating innovative education tools for NTNU teachers and students.
There are 2 possible research questions in this project:1. Development and evaluation of virtual classrooms and learning areas, populated by virtual humans/ teaching assistants powered by ChatGPT or similar chatbots (in collaboration with NTNU teachers, using existing ChatGPT plugins). These teaching assistants will be able to interact with students 24/7, answering their questions and providing assistance and personalized feedback. 2. Explore how generative AI can be used to rapidly create 3D educational content for use in such virtual classrooms by teachers/students without prior programming experience. Previous studies among teachers at NTNU and similar international studies identified difficulties with generation of educational content as one of the major obstacles to wider adoption of XR among educators.
The students will have access to a very well-equipped IMTEL VR lab (https://www.ntnu.edu/imtel/) containing Apple Vision Pro, Valve Index, HTC Vive/Vives Pros, Vive Cosmos, 2 Magic Leaps, several Hololenses 1 and 2, Mixed Reality headsets, Oculus Quests, Oculus Rifts, VR treadmill Virtuix Omni & Cyberith Virtualizer, BHaptics suit & gloves, VR laptops etc. A significant number of the VR/AR equipment is portable and can be used at home.
Supervisors: Monica Divitini, Ekaterina Prasolova-Førland (ekaterip@ntnu.no) & Mikhail Fominykh!!!PLEASE CONTACT Prof. Prasolova-Førland for more information about the task!!!
Objective: To design and integrate AI-powered chatbots into simulations to provide real-time scaffolding and analyze their effectiveness in addressing learner challenges and improving outcomes.
Description: This research focuses on embedding AI-driven chatbots into Articulate Storyline 360 simulations to provide real-time scaffolding and support for learners. The study will evaluate the effectiveness of these chatbots in addressing common learner challenges, such as navigating complex tasks or understanding difficult concepts. By analyzing learner interactions and outcomes, the research will offer insights into the potential of conversational AI to personalize and enhance online learning experiences.
In this project, the joint interests of advisor and student(s) come together.
Typically, the project will be based on previous research by Prof. Ole Jakob Mengshoel AND interests of students. If only one of these is present, there is no basis for a project.
TL;DR: The Eclipse Modeling Framework it is the main open source framework for working with Model-Driven Engineering tasks (think of TDT4250). We want to investigate what are the available alternatives and what features they support.
The Eclipse Modeling Framework (EMF) is a Java-based framework that enables the application Model-Driven Engineering (MDE) practices within Eclipse. Its introduction in the early 2000's [1] had a dramatic impact to the popularization of MDE practices and has revolutionized the way we think about and act on models, at least in its community. EMF provides the foundations for metamodeling, model transformation, and code generation tasks within Eclipse, allowing users to develop their own Domain-Specific Languages (DSLs), and creating a large part of the language infrastructure with reduced effort.
However, while Java and Eclipse were at the peak of their popularity in the early 2000's, the panorama has greatly changed today. For example, the explosion of machine learning has popularized more flexible languages like Python; continuous integration practices are pushing towards textual-based configuration languages rather than graphical models; and web-based applications are gaining advantage over desktop applications like Eclipse.
For these reasons, practical alternatives to EMF need to be found, to be able to apply MDE techniques and practices to these new development scenarios. Alternative platforms have been developed in recent years [2], including re-implementations of EMF in other languages [3]. Furthermore, other communities are incorporating similar concepts, like for example the MLIR project built around LLVM [4]. At the same time, the EMF "ecosystem" is extremely rich and diverse, and no other framework has succeeded to date to provide comparable functionality.
The goal of this project is to analyze alternatives to EMF and tools having a similar objective, understanding their similarities and differences. The tools will be exercised on standardized scenarios, which will allow a fair comparison of their features.
In more details, this project involves the following activities:
Profile
TL;DR: Evaluate the ability of Large Language Models (LLMs) to generate code in a specific programming language that is used for manipulating graphical models in software engineering. Experience with LLMs is recommended. Will use publicly available data.
Note: This project combines two research areas that use the term “model” with different meanings, please don’t get confused by that!
Model transformation is a software engineering task in which semi-formal models (e.g., UML diagrams) are modified by automated transformation. It is related to the Model-Driven Engineering area (MDE) [1], and here “model” means an abstraction of a system in some language, for example an UML or SysML diagram. Those transformations are typically written in specializaed programming languages such as ATL [3] or Epsilon [4].
Large Language Models (LLMs) [2] is a recently coined term in the field of machine learning applied to the processing of natural language. The term identifies machine learning models with a large number of parameters, typically in the scale of billions. Those models have shown impressing performance in different tasks related to manipulation of text, including generation and understanding of both natural language and source code. ChatGPT is one of the most prominent applications of LLMs. Simplifying, the term “model” here essentially means a neural nework.
Research on the use of LLMs for software engineering tasks is emerging, for example for code refactoring tasks, writing test cases, etc. The objective of this project is to investigate the use of LLMs for model-transformation tasks. Writing model transformations is a difficult task, but creating input/output pairs of models to be transformed is instead usually quite simple. The idea is to exploit the generalization abilities of LLMs to write a correct model transformation from examples.
The long-term research objective linked to this activity is to build a model transformation engine based on LLMs.
TL;DR: Apply Large Language Models (LLMs) to the problem of verifying if source code satisfies given coding rules. Coding quality rules exist for different purposes, for example the SEI Cert Coding Standards focus on security. Experience with LLMs is recommended.
Coding conventions are guidelines for software development that impose constraints on how to write source code in a certain programming language. In this project we are mostly interested in rules that enforce quality properties like security or performance. In general, the adherence to precise coding rules avoids introducing known bugs, and it is a fundamental practice for ensuring the reliability of complex software systems. Coding quality rules exist for different purposes; the SEI Cert Oracle Coding Standard for Java is a good example of a coding standard that focuses on security [1].
Some rules can be checked by static analysis tools, such as PMD [3] or SonarQube [4]. However, such tools only support a reduced number of rules. Further, coding conventions are not static artifacts; rather, they evolve over time, following the introduction of new language features or the discovery of new vulnerabilities.
Large Language Models (LLMs) [2] is a recently coined term in the field of machine learning applied to the processing of natural language. The term identifies machine learning models with a large number of parameters, typically in the scale of billions. Those models have shown impressing performance in different tasks related to manipulation of text, including generation and understanding of both natural language and source code. ChatGPT is one of the most prominent applications of LLMs.
Research on the use of LLMs for software engineering tasks is emerging, for example for code refactoring tasks, writing test cases, etc. The objective of this project is to investigate the use of LLMs for the verification of coding rules. The idea is to exploit the generalization abilities of LLMs and their ability to handle textual data to automate the process of checking whether a certain source code satisfy a coding rule specified in natural language.
The long-term research objective linked to this activity is to build a framework that can automate the implementation of checkers for new coding rules specified in natural language.
This is a project targeting selected AI-related aspects of a Situation Awareness System for an autonomous ferry, and will be performed in close cooperation with the Trondheim-based company Zeabuz.
The project builds on solid intermediate results of a previous MSc project, performed between Aug 2022 and May 2023.
"Situation Awareness" deals with the process of processing sensor data from different sensor modalities on a moving system, in the present case: an autonomous passenger ferry and building a valid dynamic representation of the environment around the regarded mobile system. Such a dynamic representation allows to navigate safely, avoid colliisions with static obstacles and other moving vehicles, and is the basis for performing complex maneuvers such as docking a ship.
The project leaves ample space for focusing on different aspects of situation awareness and AI-based processing of sensor data, depending on the interest of the student(s) and his/her/their pre-knowledge.
The work could be focusing on typical machine learning aspects, or on "classical" (in particular statistical model-based) methods.
Background and Problem Description
In spite of impressive progress in recent years, there are still limitations and opportunities related to language technology the Norwegian language. There are still significant application areas that require innovative solutions and breakthroughs in order to be fully realized. Issues that must be resolved include both language technology in general and applications for the Norwegian language (and other resource-constrained languages). In fact, the importance of improving language technology in the handling of Norwegian has been highlighted by the Norwegian government's AI strategy document.
The student should investigate language technology, with a particular focus on building upon large language models such as GPT and LLaMA. Studies of both English and Norwegian languages would be of interest. Potential application areas include dialogue generation [1], chatbots [2], and question generation from text.
This project has some flexibility for the student to pursue their own interests when it comes to the focus of the work. (1) One option is for the student(s) to work with the Norwegian bank DNB (see contact above). They have observed a recent rise of new developments in machine learning and large language models. There is consequently a large interest from business and support areas to explore how these technologies can be applied to improve and automate a wide range of internal processes at DNB. They need a framework for analysis and proof of concept to help us choose the right technologies and the best use cases to prioritize. (2) Another option is to focus on open-source data and more fundamental questions when it comes to large language models.
Tasks for the student to perform, after consulting with advisor and potential external contacts, include:
There are many relevant datasets available, both in English and Norwegian. Appropriate dataset(s) will be identified based on discussions between the student(s) and advisor(s). Here are some example datasets:
[1] Ellen Zhang Chang. Surrounding Dialogue Generation using Deep Learning with Adapters. MS Thesis, NTNU, 2022.
[2] Liu, B., Yu, T., Lane, I., and Mengshoel, O. J. (2018, April). Customized nonlinear bandits for online response selection in neural conversation models. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 32, No. 1).
Agile software development methods have gained increasing popularity in software development projecs. Agile methods prescribe practices for development, and were first used in small projects with little criticality. However, such methods are increasingly used in large projects, and this project will investigate how the practices are adapted and combined with traditional practices to function effectively in large scale. A first generation of large-scale agile methods combined advice from methods such as Scrum with advice from project management. A second generation of methods are currently taken up by the global software industry, with methods such as the Scaled Agile Framework, Large-Scale Scrum, the Spotify model and Disciplined Agile Delivery.
The project will start as a literature review, but which can be extended with an empirical study for a master thesis project. For an introduction to large-scale agile development, see introduction to special isue in the IEEE Software magazine: arxiv.org/abs/1901.00324 For an example study of a large-scale agile development project see: https://link.springer.com/article/10.1007/s10664-017-9524-2
A master thesis can be partially based on material from the Agile 2.0 competence-building project where NTNU was a partner, and with other partners DNV GL, Equinor, Kantega, Kongsberg Defence & Aerospace, Sopra Steria and Sticos. The project was led by SINTEF Digital and supported by the Research Council of Norway.
The task will start with a review about how AI is currently introduced to students at schools (lower and upper secondary), either as part of the curriculum or in extra-curricular activities. The task will then continue with the design of an intervention that can be used to introduce AI in a an engagement and playful way, with focus on responsability and sustainability.
Supervisors: Michail Giannakos Place: LCI Lab: https://lci.idi.ntnu.no/ Suitable for: One or two students
Introduction Learning analytics has been a hot topic for a while in educational communities, organizations and institutions. There are four essential elements involved in all learning analytics processes: data, analysis, report and action. Learning analytics are important because every “trace” within an electronic learning environment may be valuable information that can be tracked, analyzed and combined with external learner data; every simple or more complex action within such environments provide insights that can guide decision making (e.g., students, teachers, policy makers).
Thesis Description The increased need to inform decisions and take actions based on data, points out the significance of understanding and adopting learning analytics in everyday educational practice. And in order to treat educational data in a respectful and protected manner, the policies for learning analytics play a major role and need to be explicitly clarified. This thesis will analyse data associated with the use of learning analytics and AI learning systems in Norway, and has the option to also collect primary data (e.g., questionnaires or interviews with students and lecturers), with an ultimate goal to identify what LA and AI learning systems are used in Norway, how they are put into practice and potential challenges and opportunities with their us.
Requirements The ideal candidate will have a background and interest in data analysis and research methods, no programming skills are required.
Relevant information The ideal candidate will have a background in data analysis, no programming skills are required. - The candidate can use data that have been collected in the context of the national expert group in learning analytics: https://laringsanalyse.no/ - The candidate can use data from different national organizations as Sikt and NOKUT. - The national interim report in Learning analytics (https://www.regjeringen.no/no/dokumenter/laringsanalyse-noen-sentrale-dilemmaer/id29167 47/) and the final report (is going to be published in June), can also serve as useful sources.
See the complete topic as PDF: https://drive.google.com/file/d/1wv5l2eok3LLfTuGucurgHEJ7Z9RSztVc/view?usp=sharing
Dette informatikkdidaktiske prosjektet omhandler hvordan læreres tilnærming til undervisning påvirkes av elevenes forutsetninger for å lære programmering. Kandidaten vil utforske litteraturen rundt hvilke forutsetninger som påvirker elevers evne til å lære programmering, hvordan undervisning kan tilpasses ulike forutsetninger, og læreres bevissthet rundt dette.
Co-supervisor: Gabrielle Hansen, Excited
How do software companies ensure organisational learning at different organisatinoal levels? Learning organisations and knowledge management have been vital areas for software organisations. Recently, many organisations have focused their learning activities at team level rather than individual level. This project will first survey literature on organisational learning and knowledge management in software engineering, and as a possible extention in a master thesis conduct an empirical study in a Norwegian software engineering environment. The student can suggest a case, or a case can be found through the supervisor´s network.
In our group, we have procured a LIMO small scale floor robot from AgileXhttps://global.agilex.ai/education/4
The LIMO robot is equipped with a stereo camera (including depth image generation), a Lidar sensor, and a Jetson Nano GPU board enabling e.g. object recognition. The LIMO runs the robot operating system ROS, and in currently ongoing projects, students create the foundational AI capabilities like environment mapping, and path planning.
In the new project season H2024-V2025 the intention is to tackle new AI challenges for this robot:
For instance, we could have the LIMO seek for hidden “treasure objects” in the AI Lab, or recognize selected AI Lab visitors, approach them autonomously, and greet them, or perform “catch me if you can” with two LIMOs.
Many kinds of interesting and demanding AI robotics tasks can be addressed, and it is explicitly possible that students contribute their own ideas of what one or several LIMOs can do, including also speech interaction.
Develop a simple Computerized adaptive testing (CAT) that operates with multiple choice (MC) questions and provides explanations based on a conversational agent/LLM. In other words, a teacher can load MC questions into a system and create a test, and allow students to log in and test their skills on a specific topic. The system will be adaptive, in a way that it will give you questions that are appropriate to your skills (e.g. if you fail a medium-difficulty question it moves you to an easier one, if you are correct, it moves you to a more difficult one). The system will be tested using this question bank: https://web-cat.org/questionbank/ The goal of this topic is to investigate and evaluate the use of CAT in CS1 and CS2 topics in HE.
Læringsteknologi er programvare og andre teknologiske produkter som understøtter læring og undervisning. Her er det mulighet for selvvalgte oppgaver enten fra studenter eller studenter i samarbeid med fagstab, og prosjekter som kan relateres til enten Excited senter for fremragende utdanning.
In this collaborative initiative we are seeking to enable the development and deployment of autonomous marine vessels. The autonomous technology company Zeabuz (POC Dr. Smogeli), which is located in Trondheim, is developing autonomous urban passenger ferries and seeking to perform state-of-the-art safety validation. Safety is also a key challenge in aerospace, and the concepts of adaptive stress testing has been developed by NASA Ames Research Center (ARC) and NASA’s partners to meet that challenge. At NTNU, the Norwegian Open AI Lab (advisor Prof. Mengshoel) will be advising the student(s) working on the project.
It is forecast that a steadily increasing percentage of the world population will be living in cities. However, city growth is often limited by lack of scalable infrastructure and congested road traffic leading to excessive emissions. At the same time, many cities are built by historical waterways such as lakes, rivers, canals, bays, fjords or harbour basins. Often, these waterways are largely unused, while they separate districts and limit mobility. Meeting these challenges with bridges and tunnels is a costly, non-scalable and inflexible solution, causing a large footprint while not meeting future problems due to likely increasing water levels. In contrast, establishing shortcuts across and along existing waterways that connect and revitalize urban areas is a sustainable, flexible, efficient and compelling solution for cities and citizens.
Zeabuz is developing core technologies and solutions that can enable more environmentally friendly and flexible mobility solutions in urban areas. The Zeabuz mobility system consists of a network of electric, autonomous passenger ferries with centrally located docking and charging stations, high bandwidth communication and a remote support centre.
In 2022 NTNU and Zeabuz together demonstrated the world’s first autonomous urban passenger ferry – the milliAmpere 2 (mA2) – in Trondheim. The mA2 is a fully operational passenger ferry with an autonomy system developed by Zeabuz, designed to operate along a fixed path across the canal between Ravnkloa and Vestre Kanalkai. The autonomy system is able to avoid objects and handle any traffic situation that could occur in this Operational Design Domain (ODD).
In parallel, NTNU and Zeabuz have developed an mA2 digital twin, which is a complete simulator of the autonomous ferry and its operating environment in the Trondheim canal. This enables simulation of any kind of operational scenario in a synthetic environment.
The Zeabuz autonomy system is a complex, software intensive system subject to an unpredictable operating environment, that needs to be actively managed throughout its life-cycle. This makes formal safety proofs (practically) impossible. Instead, one needs to resort to statistical considerations in the safety argumentation. In other words, there is a need to argue – based on the accumulated experience, testing, verification and validation activities we have collected – that the system is sufficiently safe.
Experience from the aerospace and automotive industries, which have been dealing with similar challenges for autonomous aircraft and cars for some time, shows that testing and real-world experience is not sufficient – even with a large fleet of aircraft or cars. The reason for this is that the number of possible critical scenarios the system needs to handle is practically unlimited. The only viable solution appears to be large-scale simulations of the system and its environment, where the full parameter space of possible scenarios in principle can be explored [2]. However, this triggers the problem of designing and selecting which scenarios to run among the near infinite number of possible parameter combinations. Example questions that show up are: How to ensure that scenarios are relevant and representative? How to cover the critical “corner cases”? How to find the “weak spots” in the system and force it towards failure? How to evaluate the system performance?
To solve these issues, we need a systematic and effective way of designing, running, and evaluating simulation scenarios that together give sufficient confidence in the safety of the Zeabuz autonomous ferry system. This proposed MS project will be contributing towards this goal.
The project is suitable for one or two students (for simplicity we will say “students” below). The students will get access to the milliAmpere 2 digital twin, which is a complete simulator of the autonomous ferry and its operating environment in the Trondheim canal. For this project, the simulations can be parameterized in terms of the traffic situation, meaning the candidate can specify what kind of vessels that will be part of the scene and how they will move and act with respect to mA2.
In principle, any kind of operational scenario can be designed and deployed in the mA2 digital twin, meaning it can serve as the backbone of such a test system.
Depending on actual data availability, the candidate may also get access to full-scale data from experiments, if this is found to be of relevance.
The thesis will employ, adapt and extend a method called adaptive stress testing (AST) to find likely failure events of the autonomy system in mA2, using reinforcement learning (RL) and other AI and ML techniques [1,2,3,4,5]. The AST method has previously been used to validate aircraft collision avoidance systems [3,4,5], which is a similar problem. There are initial results for mA2 [6].
A follow-up MS Thesis will very much depend on the fall project. This is a sketch for one possible follow-up MS Thesis:
Regarding «algorithmic extensions to AST» as mentioned above: Since AST is based on RL, exploration of these ideas will first start on the RL side, and then be adapted to AST based on experiences with the Zeabuz requirements and simulation. Here are some ideas for such algorithmic extensions:
If an extension project is only partially completed, contributions can still be made to RL.
We foresee no business or scientific restrictions for the project. The candidate will have unrestricted access to running simulations from the Zeabuz side. Computational resources are made available by NTNU. The autonomy team at Zeabuz in Trondheim will be available for support and guidance regarding interfacing and other practicalities.
[1] Browne, C. B., Powley, E., Whitehouse, D., Lucas, S. M., Cowling, P. I., Rohlfshagen, P., Tavener, S., Perez, D., Samothrakis, S., & Colton, S. (2012). A survey of Monte Carlo tree search methods. IEEE Transactions on Computational Intelligence and AI in Games, 4(1), 1–43.
[2] A. Corso, R. J. Moss, M. Koren, R. Lee, and M. J. Kochenderfer, “A Survey of Algorithms for Black-box Safety Validation,” ArXiv e-prints, iss. 2005.02979, 2020.
[3] R. J. Moss, R. Lee, and M. J. Kochenderfer, “Adaptive Stress Testing of Trajectory Predictions in Flight Management Systems,” in IEEE/AIAA Digital Avionics Systems Conference (DASC), 2020.
[4] R. Lee, O. J. Mengshoel, Saksena Anshu, R. Gardner, D. Genin, J. Silbermann, M. Owen, and M. J. Kochenderfer, “Adaptive Stress Testing: Finding Likely Failure Events with Reinforcement Learning,” Journal of Artificial Intelligence Research, vol. 69, p. 1165–1201, 2020.
[5] Lee, R., Kochenderfer, M. J., Mengshoel, O. J., Brat, G. P., & Owen, M. P. (2015). Adaptive stress testing of airborne collision avoidance systems. In Digital Avionics Systems Conference (DASC). AIAA/IEEE.
[6] Hjelmeland, H. W., Eriksen, B. O. H., Mengshoel, O. J., & Lekkas, A. M. (2022). Identification of Failure Modes in the Collision Avoidance System of an Autonomous Ferry using Adaptive Stress Testing. IFAC-PapersOnLine, 55(31), 470-477.
At SINTEF Ocean we are interested in applying machine learning to ocean data to find patterns or neighbourhoods, i.e., areas or periods with specific properties that are interesting or relevant for marine industries, as they determine fish health, transport of pollution and more.
At SINTEF, we are often asked to assess ocean locations with respect to environmental properties or marine activities with respect to environmental risk. In order to do that, it is preferable to have several years with modelled ocean data to account for different seasons and yearly variations in conditions. It is hard for a human to understand these data and to pick periods or areas that are relevant or typical for the application. Per today these data are analysed with normal statistics, like means / max / variations. For operations planning or planned discharges it would be nice to understand patterns in the data and either find suitable periods or areas in the ocean or be able to exclude them for protection.
Machine learning for effective ocean data analysis Apply different machine learning methods to find commonalities in ocean data that are meaningful to marine operations (pollution transport, aquaculture). Ocean data is 4 dimensional, with 3 spatial dimensions plus time, so the data is like several sensor readings (currents, temperature, salinity) over time and in space (like cubes in a Rubik cube, with each cube being several sensors). We want to understand the correlation and patterns in these data.
We can use data that is available from met via https://thredds.met.no/ and possibly combine modelled (Ocean & Ice, Norkyst800) with observations (Observations, SVV E39) and aquaculture production data from Barentswatch https://www.barentswatch.no/havbruk/? Ocean model data come in netCDF and are around 5 GB per month for whole Norwegian waters. A subset will be sufficient for this work.
Potential Challenges
The proposed approach has so far been applied to greater ocean areas or features like daily or monthly means. These kind of features are unsuitable for environmental risk assessment or operation planning. The application to local environments is new and challenging.
Integrating multimodal data, specifically visual and language data, is the focus of this project. Much of the progress has been due to advances in machine learning and the availability of suitable data sets consisting of images, each with multiple captions.
Better understanding the integratiion of visual and language intelligence in humans is fundamentally important in cognitive science and also has more short-term applications. One example of an important application is to improve accessibility for the visually impaired via creation of image captions, for which speech generation is possible.
We describe two potential directions that the project can take below, namely (a) machine learning for image captioning and (b) machine learning for multimodal named entity recognition.
(a) Machine Learning for Image Captioning
Image captioning, a challenging but exciting area of research at the intersection of natural language processing and computer vision, has seen dramatic progress over the last few years. However, to the best of our knowledge, there has been little research done on the impact of data augmentation on the quality of image captioning results. Progress has, though, been made on the following question [1,2]: does it improve image captioning performance if we increase the number of captions in a dataset by augmenting the existing captions with paraphrases?
In this project, we will address this question and build on existing work performed at NTNU and CMU [1,2,3], which has made several contributions including the following. First, we presented a novel method of adding captions to a dataset by means of paraphrasing. Second, using our method, we generated a set of paraphrases for the seminal MS COCO dataset. Our paraphrase dataset consists of around 490 000 paraphrased captions that are different from the original captions in MS COCO. Third, in image captioning experiments, using the novel paraphrase dataset, we presented machine learning results using deep learning. These results clearly suggesed that the process of extending image captioning training datasets with paraphrases improves image captioning models' performance.
Proposed areas of work for a project and MS-thesis are to: (1) research alternative machine learning models, including models found via network architecture search; (2) study alternative image and word embeddings, including pre-trained embeddings; (3) consider alternative ways of doing data augmentation, including augmentation via different paraphrasing methods; and (4) study other datasets (including Norwegian language data) and (5) develop improved metrics for image captioning.
Regarding Point (4) above, two ideas are to use NRK or NTB data. For example, NTB’s picture archive consistsing of over 80 million images. Overall these images include well structured metadata that can be used as input for the project. Existing AI will form the basis and our face and object detection will be important steps in further AI development. The students will be able to use Amazon’s recognition tools and existing models for developing new models. NTB's 80million images are richly and thoroughly tagged with metadata. Those very well-tagged images are the result of many years of manual tagging work. All NTB images and videos contain many metadata fields like: Title, Terms, Keyword, Persons, Description, Date, Places, Gps etc.
[1] I. R. Turkerud. "Image Captioning with Deep Learning." Master’s thesis. NTNU, Trondheim, Norway, 2020.
[2] I. R. Turkerud and O. J. Mengshoel, "Image Captioning using Deep Learning: Text Augmentation by Paraphrasing via Backtranslation," 2021 IEEE Symposium Series on Computational Intelligence (SSCI), 2021, pp. 01-10, doi: 10.1109/SSCI50451.2021.9659834.
[3] A. Rajendra, R. Rajendra, O. J. Mengshoel, M. Zeng and M. Haider. "Captioning with Language-Based Attention." 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA), 2018, pp. 415-423, doi: 10.1109/DSAA.2018.00054.
(b) Machine Learning for Multimodal Named Entity Recognition
Another related task is multimodal named entity recognition. Named entity recognition (NER) is an important task in NLP, where we extract entities such as person and location from textual data. Previous NER methods usually only used the textual content. However, many modern datasets contain not only textual content, but also images. In multimodal NER, it is interesting to study how to effectively combine the textual content and images, to better understand the text and extract named entities.
Example data and source code related to multimodal named entity recognition: https://github.com/jinlanfu/NERmultimodal
[1] Zhang, Qi, et al. "Adaptive co-attention network for named entity recognition in tweets." Thirty-Second AAAI Conference on Artificial Intelligence. 2018.
[2] Lu, Di, et al. "Visual attention model for name tagging in multimodal social media." Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2018.
[3] Sun, Lin, et al. "RIVA: a pre-trained tweet multimodal model based on text-image relation for multimodal NER." Proceedings of the 28th International Conference on Computational Linguistics. 2020
The goal of this project is to apply techniques from machine learning to optimize the design of nanomagnetic computing systems. This is an interdisciplinary project at the intersection between computing, artificial intelligence and physics.
A key challenge in ASI computing systems is optimizing the placement and orientation of the magnets such that some desired functionality is obtained. With thousands of magnets, the parameter space is way too big for exhaustive search methods. However, modern machine learning methods are routinely applied to optimize neural networks with billions of parameters. Frameworks such as PyTorch makes these methods readily applicable to non-neural systems. For instance, PyTorch was recently employed to optimize a nanomagnet-based spin-wave scatterer to perform computing tasks such as vowel recognition. This was made possible by incorporating PyTorch gradients into the physical simulator itself.
Within COMET we are exploring computing in ASI systems. Towards that end we have developed a large-scale ASI simulator called flatspin. Integrating PyTorch (or similar) into flatspin should open for the possibility to optimize ASI designs towards some target function. Alternatively, other machine methods such as evolutionary search may also be relevant.
Artificial Spin Ice (ASI) are metamaterial systems consisting of coupled nano magnets arranged on a 2D lattice, whose collective large-scale emergent behavior has attracted considerable interest as computing substrates.
The goal is to develop a machine learning-based tool that can predict the revenue of an actor in a MaaS ecosystem with predicted ticket sales.
Context: With Mobility-as-a-Service (MaaS), a diversity of transport services is offered as one integrated service to travellers via one App and with one payment for the whole journey. A MaaS service may involve many transport service providers with different transport services/modes, for example, public transport (bus, ferry and train), on demand services (taxi, city bike, scooter) and shared cars. The goal of MaaS is more sustainable transport by providing an easy and flexible mobility service that can replace private cars.
Problem: The lack of viable business models has been a barrier for the establishment of MaaS. It must be profitable both for the provider of the MaaS service and the transport service providers. The motivation for a transport service provider to join a MaaS ecosystem is in most cases the opportunity to reach new customer groups and bigger markets. It is, however, difficult to predict this in advance.
Requirement: A tool that can predict the economic effects of a MaaS service for all participants, both the MaaS provider and the transport service providers. A forecast on the revenue (expected ticket sales) would help a transport service provider to decide whether to join an existing MaaS ecosystem.
Task: The task is to build a machine learning model to predict the revenue of participants in the MaaS ecosystem by exploring relevant data that may affect the ticket sales, such as the travel patterns in the region covered, the demographic data of the population in the region, weather data, parking information and others.
The work will be in collaboration with SINTEF Digital, done in the context of a research project MaaSeKopp, where a number of data collected in the project could be used as input to the machine learning model. In addition, relevant open data will be used.
Project Background
This project is in the area of artificial intelligence for stochastic optimization. The emphasis will be on the integration of methods from evolutionary algorithms and stochastic local search, with a focus on both theory and applications.
In artificial intelligence and machine learning, there are several methods that use or rely on randomization: mutation and crossover in evolutionary algorithms [Goldberg, 1989], probabilistic acceptance in simulated annealing, dropout and stochastic gradient descent in deep learning, stochastic local search (SLS) [Selman et al., 1992] [Hoos & Stützle, 2005] [Mengshoel et al., 1998], and randomization in systematic search [Gomes et al., 1998].
Stochastic optimization (SO) algorithms, which include stochastic local search (SLS) and evolutionary algorithms (EAs), will be a main focus in this project. Stochastic optimization (SO) algorithms are among the best in solving computationally hard problems including NP-hard problems such as satisfiability (SAT), traveling salesman (TSP), scheduling, and most probable explanations in Bayesian networks (BNs) [Selman et al., 1992] [Hoos & Stützle, 2005] [Mengshoel et al., 2010] [Mengshoel et al., 2016] [Pal & Mengshoel, 2016].
Research on SO performed by me and my collaborators has focused on (i) improving the theoretical foundation of SO; (ii) empirically demonstrating the benefit of SO using natural and synthetic problem instances; and (iiii) applying the approach in diverse areas of artificial intelligence and machine learning. For (i), we have for instance analyzed homogenous Markov chains to derive instructive expected hitting time results [Mengshoel, 2008] [Mengshoel et al., 2016]. For (ii) and (iii), we have demonstrated that SO algorithms can be competitive for applications including computing the most probably explanation (MPE) in Bayesian networks; feature selection in machine learning; and sparse signal recovery in signal processing [Mengshoel et al., 2010] [Mengshoel et al., 2011] [Mengshoel et al., 2014] [Pal & Mengshoel, 2016] [Yu et al., 2017] [Liu et al., 2018].
Despite several successes, the current state-of-the-art suffers from some drawbacks, creating research opportunities in several areas. In this research project, we will explore those opportunities and develop novel ideas at the interface of stochastic local search, machine learning, evolutionary algorithms, probabilistic graphical models, and deep learning. Special attention will be paid to multimodal functions, which play a central role in artificial intelligence and evolutionary algorithms. There appears to be several limiting factors when it comes to existing work on SO for such multimodal functions, including the following. Firsrt, multimodal optimization algorithms often search for “as many local optima as possible” without input or guidance from a decision maker. Second, evolutionary algorithms have been used for certain combinatorial optimization problems including feature selection, but there has been surprisingly little emphasis on diversity maintenance including niching in research on machine learning including feature selection.
The integration of local search and evolutionary algorithms has, to some extent, been studied in the literature, often under the term memetic algorithms (MAs) [Eiben & Smith, 2015] [Krasnogor & Smith, 2005] [Xue et al., 2016]. Intuitively, the goal of an MA is for the EA component to search well globally, while the SLS component’s goal is to optimize locally. In other words, one wants “the best of both worlds.” There is a broad diversity of different MA types and designs [Krasnogor & Smith, 2005] [Xue et al., 2016]. Both Lamarckian and Baldwinian MAs have been studied, reflecting the spirited discussion between Lamarck and Baldwin in evolutionary biology in the late 1800. Current research seems to favor Lamarckian MAs, in which the locally optimized individual replaces the original individual in the population.
Focus in Project
The focus of the project is on the integration of SLS and EA methods, inspired by so-called memetic algorithms (MAs), to be analyzed and tested on synthetic and natural problems. Several MA-based approaches to feature selection [Kannan & Ramaraj, 2010] [Xue et al., 2016] [Lee et al., 2019] have been developed, inspiring this proposed research project.
To be investigated are methods that addresses the issues identified above, and perhaps others, by integrating multiple different methods, and carefully balancing their execution via parameter tuning and control [Krafotias et al., 2015] [Mengshoel et al., 2014] [Mengshoel et al., 2016] [Yu et al., 2017]. An interesting idea is to combine stochastic local search, evolutionary computing, clustering, and feedback control algorithms, with emphasis on the balance between exploration and exploitation during search. Such balance can be achieved by means of niching, for example crowding [Mengshoel et al., 2014]. The goal will be to empirically test such a multi-method algorithm on synthetic and natural combinatorial function optimization problems, including ML problems (for example feature selection or network search problems), and empirically establish performance relative to other methods. For the more theoretically oriented student, theory development would be of great interest [Mengshoel, 2008] [Nguyen & Sudholt, 2020].
The project can be suitable for one or two students. The students will participate in one or more of the following tasks in this proposed project:
We already have implementations of MA, SLS, and EA methods that can be used as starting points in the project. We also have concrete ideas about problems and datasets to be considered, based on the interest of the student or students.
[Eiben & Smith, 2015] A. E. Eiben and J. E. Smith. 2015. Introduction to Evolutionary Computing (2nd ed.). Springer-Verlag Berlin Heidelberg, Chapter 9.1.
[Elsken et al., 2019] T. Elsken, J. H. Metzen, and F. Hutter. Neural Architecture Search: A Survey. Journal of Machine Learning Research, Vol. 20, Iss. 55, 2019.
[Goldberg, 1989] D. E. Goldberg. Genetic Algorithms in Search, Optimization and Machine Learning (1st ed.). Addison-Wesley, Boston, MA, 1989.
[Gomes et al., 1998] C.P. Gomes, B. Selman, and H. Kautz, Boosting combinatorial search through randomization, in: Proc. of the Fifteenth National Conference on Artificial Intelligence (AAAI-98), Madison, WI, 1998, pp. 431–437.
[Hoos & Stützle, 2005] H.H. Hoos and T. Stützle, Stochastic Local Search: Foundations and Applications, Morgan Kaufmann, San Francisco, CA, 2005.
[Kannan & Ramaraj, 2010] S. S. Kannan and N. Ramaraj. A novel hybrid feature selection via Symmetrical Uncertainty ranking based local memetic search algorithm. Knowledge-Based Systems 23, 6 (2010), pp. 580–585. https://doi.org/10.1016/j.knosys.2010.03.016
[Krafotias et al., 2015] G. Krafotias, M. Hoogendoorn, and A. E. Eiben. 2015. Parameter Control in Evolutionary Algorithms: Trends and Challenges. IEEE Transactions on Evolutionary Computation 19, 2 (2015), pp. 167–187.
[Krasnogor & Smith, 2005] N. Krasnogor and J. Smith. 2005. A tutorial for competent memetic algorithms: model, taxonomy, and design issues. IEEE Transactions on Evolutionary Computation 9, 5 (2005), pp. 474–488.
[Lee et al., 2019] [24] J. Lee, I. Yu, J. Park, and D.-W. Kim. Memetic feature selection for multilabel text categorization using label frequency difference. Information Sciences 485 (2019), pp. 263 – 280.
[Liu et al., 2018] B. Liu, T. Yu, I. Lane, and O. J. Mengshoel. Customized Nonlinear Bandits for Online Response Selection in Neural Conversation Models, in Proc. AAAI 2018.
[Mengshoel, 2008] O. J. Mengshoel. Understanding the Role of Noise in Stochastic Local Search: Analysis and Experiments. Artificial Intelligence Vol. 172, Iss. 8-9, 2008.
[Mengshoel et al., 2010] O. J. Mengshoel, D. Roth, and D. C. Wilkins. Portfolios in Stochastic Local Search: Efficiently Computing Most Probable Explanations in Bayesian Networks. J. Autom. Reasoning, 2010.
[Mengshoel et al., 2011] O. J. Mengshoel, D. C. Wilkins, and D. Roth. Initialization and Restart in Stochastic Local Search: Computing a Most Probable Explanation in Bayesian Networks. IEEE Transactions on Knowledge and Data Engineering, Vol. 23, Iss. 2, 2011.
[Mengshoel et al., 2014] O. J. Mengshoel, S. F. Galán, and A. De Dios. 2014. Adaptive generalized crowding for genetic algorithms. Inf. Sci. 258 (February, 2014), pp. 140–159. DOI: https://doi.org/10.1016/j.ins.2013.08.056.
[Mengshoel et al., 2016] O. J. Mengshoel, Y. Ahres, and T. Yu. Markov Chain Analysis of Noise and Restart in Stochastic Local Search, in: Proc. of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16), 2016.
[Nguyen & Sudholt, 2020] P. T. H. Nguyen and D. Sudholt. Memetic algorithms outperform evolutionary algorithms in multimodal optimisation, Artificial Intelligence, Volume 287 (2020), pp. 1 - 21.
[Pal & Mengshoel, 2016] D. K. Pal and O. J. Mengshoel. Stochastic CoSaMP: Randomizing Greedy Pursuit for Sparse Signal Recovery, in: Proc. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery, p. 761 – 776, 2016.
[Selman et al., 1992] B. Selman, H. Levesque, and D. Mitchell. A new method for solving hard satisfiability problems, in: Proc. of the Tenth National Conference on Artificial Intelligence (AAAI-92), San Jose, CA, pp. 440–446, 1992.
[Xue et al., 2016] [45] B. Xue, M. Zhang, W. N. Browne, and X. Yao. 2016. A Survey on Evolutionary Computation Approaches to Feature Selection. IEEE Transactions on Evolutionary Computation 20, 4 (2016), pp. 606–626.
[Yu et al., 2017] T. Yu, B. Kveton, and O. J. Mengshoel. Thompson Sampling for Optimizing Stochastic Local Search, in: Proc. ECML PKDD, 2017.
Together with MIRA, SINTEF Health, the medical faculty at NTNU and St Olav university hospital we are offering medical image computing (MIC) projects, based on Deep Learning (DL) and Computer Vision (CV), and related to:
Some concrete examples:
It's also possible to focus more on model dev. for MIC, e.g. Vision Transformers (ViT), that can be tested on various organs and modalities.
Challenges related to MIC at key conferences like MICCAI and CVPR etc. can also be converted into interesting project proposals:
It's desirable to use MONAI (and MONAI Label) for all developments.
Main investigators for MIC-related projects are Frank Lindseth and Gabriel Kiss
Generic 3D medical image segmentation with modern neural networks that are applicable to various clinical tasks, such as segmentation of aorta in CT images, brain segmentation in MRI slices, prostate segmentation in MRI volumes.
In the age of advanced computer vision models, ensuring privacy and security is paramount. This project dives into the realm of membership inference attacks in the context of computer vision. A membership inference attack is a sophisticated form of privacy breach where an adversary attempts to determine whether a specific data point was used in training a machine learning model.
This research aims to uncover the vulnerabilities of state-of-the-art computer vision models and design defenses toward such invasive attacks. By unraveling the intricacies of this growing threat, the study contributes to the safeguarding of sensitive image data and fortifying the trustworthiness of modern computer vision systems.
NINA contributed the map of nature loss in Norway for the NRK article Norge i rødt hvitt og GRÅTT. This map was produced from outputs of a FCNN produced by Google called Dynamic World. There are two major points for improvement on the current nature loss map: (1) building a local machine learning model trained on Norwegian reference data; and (2) instead of mapping land cover for separate points in time and then analyzing the change (like in Dynamic World), create a model which directly detects and maps change. This Master's project offers a unique challenge in the realm of machine learning, diverging significantly from traditional computer vision tasks. In contrast to regular RGB images, satellite image time series provide rich spatial, multi-spectral and multi-temporal dimensions. The student might explore cutting-edge models like 3D CNNs, RNNs (including LSTMs), and Transformer models, aiming to capture the dynamic nature of ecological transformations. NINA has, through expert annotation and crowdsourcing, gathered over 10.000 polygons defining the loss of nature over Norway in the past 5 years. We also have established pre-processing pipelines for the relevant satellite imagery to be used. The Master’s student will be able to focus on testing different model architectures to leverage the multiple dimensions of the input data. The outcome will contribute substantially to environmental conservation efforts and offer a novel perspective in applying AI to ecological monitoring.
Co-advisor: Zander Venter, NINA Oslo, zander.venter@nina.no
Personality is a set of traits and unique characteristics that give both consistency and individuality to a person's behavior. As personality is accepted as an indicator of job performance, recruiters aim to retrieve these behavior traits in the screening process. One issue is that using personality questionnaires is less favored by applicants and negatively affects the pace of the recruitment process. Many recent studies started exploring asynchronous video interviews (AVI) and social platforms to predict one's personality. This study aims to explore and develop machine learning algorithms (preferably multimodal DNN) for analyzing recording interviews and accompanying resources (social media/online profile presence such as LinkedIn) in predicting one's personality on Big Five personality traits.
Game development is a large well-known area in traditional web development. However, it is still to be seen how the emerging web3 technology will take it a step further!
In this project, a multiplayer game will be developed using a mix of traditional web and modern web3 technologies! It will be an excellent opportunity for students to learn about new technologies and possibly apply those later in their thesis or in their further careers.
The game that will be developed will be a multiplayer Blackjack game unless there is a better proposal from students! Technically, it will have following components
The game will have following basic functionalities
Research Aspects
The project will mainly be based on simulations in our open-source flatspin simulator for large-scale ASI systems.
Music presents a unique way to view a dynamical system. If we can express a system's behaviour through music, then we have a new way to perceive and understand the system, as well as a novel method to create new music.
Artificial Spin Ice (ASI) are dynamical systems with a rich variety of emergent behavior. Recent work in the COMET group has demonstrated how the dynamics of ASI systems can be used to create music. First, an ASI is simulated and perturbed in some manner. Then the results are fed through a custom mapping to produce music. We also used evolutionary methods to search for an ASI geometry that produced the “best” music.
The aim of this project is to extending this work. This could be improvements to the mapping, allowing it to produce better sounding music, or music that more closely expresses the behaviour of the ASI. Alternatively, one could test out and develop different fitness functions, i.e., means of evaluating how good the produced music is, and the use this for an evolutionary search to find an ASI geometry with even better musical qualities.
Goals: The primary goal is to enhance the NBA viewing experience for fans who cannot watch games live. By developing an "Excitement Score" feature that rates games on a scale from 1 to 100, viewers will be able to select replays without spoilers based on the predicted entertainment value of the games.
1. Research and identify key metrics that significantly impact the excitement level of NBA games, such as in-game win probability, player performances, and matchup importance.
2. Collaborate with a team of developers and sports analysts from Sportradar to design an algorithm that computes the Excitement Score using real-time data and historical statistics.
3. Build the Excitement Score pipeline and make it accessible for other internal Sportradar teams, allowing for potential content teams to efficiently integrate the Excitement Score into existing applications.
4. Test the feature with a focus group of NBA fans to gather feedback and make necessary adjustments, ensuring the accuracy and relevance of the excitement scores.
Learning Objectives: Throughout this project, Sebastian will deepen his understanding of data analytics and machine learning within the sports domain. He will enhance his skills in developing AI pipelines, system integration, and user experience design. Additionally, working closely with sports analysts and data scientists from Sportradar will provide valuable insights into real-world applications of data-driven decision making. This project will also offer experience in managing a full cycle development project from conceptualization to deployment.
Interested in the future of real-time photorealistic reconstruction of complex 3D scenes?
In that case this is an excellent opportunity to understand, experiment with and explore the technology, for example in the context of autonomous driving and simulators.
Some references:
More info here
Neural Radiance Fields (NeRF) [Mildenhall et al. 2020] is a state of the art technique to generate 3D models by essentially overfitting neural networks, using them as efficient storage containers of the scene information. The NeRF method is a very simple and elegant solution to the problem of 3D reconstruction of a scene, using rudimentary ray tracing techniques for comparison with ground truth images to train the model on a basic multi-layer perceptron network. Later works have expanded this to work on random image sets [Brualla et al. 2021], for dynamic scenes [Pumarola et al. 2021], and even for scenes of any scale [Tancik et al. 2022].However, up until recently they have been quite slow methods, taking up to 2 days to train on a high-end GPU, and 1-2 minutes to reconstruct a single image of the scene. A paper just released by NVIDIA [Muller et al. 2022] changes this, going from days of training time to seconds, making real time implementation of NeRF possible.To follow up on that work, we are interested in figuring out if it is possible to apply NeRF as a tool for 3D reconstruction in real time simultaneous localization and mapping problems, producing 3D environments of real time videos.[Mildenhall et al. 2020]https://www.matthewtancik.com/nerf[Brualla et al. 2021]https://nerf-w.github.io/[Pumarola et al. 2021]https://www.albertpumarola.com/research/D-NeRF/index.html[Tancik et al. 2022]https://waymo.com/research/block-nerf/[Muller et al. 2022]https://nvlabs.github.io/instant-ngp/
This project aims to modernize an existing custom reporting service for a JavaScript-based project management application, addressing current limitations in efficiency, flexibility, and user experience. The new next-generation report service will transition from the outdated .NET and Crystal Reports-based system to a more dynamic solution utilizing API-based data extraction with JSON structures. Key features include multi-format report generation (.docx and .pdf), user-friendly report blueprint creation incorporating advanced graphical elements and 3D visualizations, large-scale data handling capabilities, and automated report generation with multi-recipient delivery. By implementing these improvements, the project seeks to streamline the reporting process, enhance data presentation and analysis capabilities, and provide a more scalable and future-proof solution that aligns with modern technological standards and user expectations in project management.
Co-supervisors:
Computed tomography (CT) is an imaging technique used to reconstruct an object’s internal structure in a non-destructive way. In medicine, CT, alongside MRI, PET, and ultrasound, is used to create 3D models of a patient that can be inspected for injuries or other changes. CT is also used in earth science and material science to better understand the internal structure of the Earth and different materials. The goal of this project is to investigate improved methods for 3D and 4D reconstruction of CT-scanned objects by means of machine learning. A recent machine learning method known as implicit neural representation will be among the methods studied. We have found in initial experiments that implicit neural representation produces state-of-the-art results, surpassing the reconstruction quality of other methods both visually and quantitatively.
The project will build on existing research performed. The research project can go in different directions based on the interests of the project team, including new students, and results achieved at the end of the Spring 2024 semester.
A team consisting of two students with partially overlapping, partially different strengts works well. One student may have a strength more towards physics and engineering, while the other student may have a strength more towards artificial intelligence and machine learning.
Ultimate Visual Computing and AI project.
Several sub-project, some examples:
We want to build a close to real-time photorealistic digital-twin of the area around the Gløs campus on top of Omniverse based on USD (or Unity / UE), as a start - can easely be scaled. The work has already started, useing existing data like terrain models, orthophotos, buildings, geo-located point-clouds and images for context, then update the model with real-time data to match the physical twin. Many of the tasks needed can be automated, for example 3D content creation from 2D images.
It should be possible to import the generated environment / model into simulators like DRIVE Sim, Isaac Sim and Carla and learn agents to drive itself (modular and end-to-end approaches like imitation and reinforment learning).
At the end we want to fine-tune the AI-models created in simulation with real data and apply the methods to our new AV platform for research on autonomous driving (AD).
Sounds interesting? Want to contribute? Let's discuss a good project fitting to your background.
Large language models (LLMs) will be gradually implemented on edge systems, like mobiles, self-driving cars, etc., as a powerful AI assistant. Normally, LLMs are trained to provide broad knowledge to different users. However, as learned from lectures on convolutional neural networks (CNNs), personalizing CNNs (removing some classes never used) can reduce the model complexity and latency while in some cases improving accuracy. The goal of this project is to evaluate an on-device personalization method for LLMs on edge systems. This means that the entire personalization procedure will be conducted on edge systems without the assistance of remote servers. Thus, the proposed method should be low overhead and hardware-aware so that the limited computational power of edge systems can be fully utilized to achieve this goal.
Denne oppgaven tilbys i samarbeid med MIA Health
I dag er MIA stort sett aktivitet (i form av puls og PAI) knyttet til forebygging, ink. muligheten for å koble seg til ulike wearables som smart-klokker og helse-monitorer, og det som trengs av frontend/backend og user-adm for å håndtere dette.
Kunne du tenke deg å være en del av et dedikert team som jobber i retning av en tvilling som passer på deg og hjelper deg gjennom livet? Noen forslag til problemstillinger som det går an å se på:
Forslagene over er akkurat det, dvs. forslag. Og kompetanse fra de fleste av IDI sine studieretninger vil være av interesse i denne konteksten. I et konkret prosjekt samarbeid etterstreber vi alltid at studenten er en del av diskusjonen og en del av forprosjektet vil bestå i å sammen spikre en master-oppgave som studenten virkelig er motivert til å jobbe med. Hvis ønskelig kan vi også avtale et møte med MIA før deadline for prosjekt ønsker går ut.
Open data involves the pooling and collecting of data across a community, industry or group of stakeholders. The motivation is the vision (aspiration, hope, belief...) that by making data openly availble, hence accessible to everyone, this will boost productivity through enhanced collaboration or create more well-functioning markets. Examples include: Open Target in pharmaceutical industry, the EU's PSD2 regulative towards open banking in finance, or HUNT research database at NTNU.
Visions of the role of open data to are widespread as illustrated by this recent Stortingsmelding, https://www.regjeringen.no/no/dokumenter/meld.-st.-22-20202021/id2841118/?ch=5
The challenge, however, is that the mere availability of open data is not sufficient for its uptake and use towards collaboration. There are social, practical and institutional conditions that need to be in place for visions of open data to materialize.
The student(s) will analyse this for a particular proposal for open data, Open Data Subsurface Universe (https://osduforum.org/). This is a data platform for sharing, communicating and doing analytics of data. It orgininated and has a foothold in the fossil energy sector, but is moving into renewable energy and CCS installations too as OSDU is a general framework for capturing any physical, geo-located asset (similar to Digital Twins).
The students will do their projects with partner companies. Presently there are two: Equinor and AkerBP.
Open data (together with data platforms and data spaces) involves collecting and sharing data across industries. The motivation is that by making data openly available, productivity is increased through enhanced collaboration or create more well-functioning markets. One successful example is how the pharmaceutical industry collaborated to develop and deploy Covid vaccines at record speed. Other examples include: the EU's PSD2 regulative towards open banking in finance and BarentsWatch for monitoring the seas.
Open data is key to the digitalization strategy by the Norwegian government as illustrated by this recent Stortingsmelding, https://www.regjeringen.no/no/dokumenter/meld.-st.-22-20202021/id2841118/?ch=5
The challenge is that there currently are social, technical, practical and organizational conditions that need to be in place for visions of open data to emerge in practice. Example challenges are motivation, feasibility of technical infrastructure, rewards and payment models.
The particular case at hand here is the Open Data Subsurface Universe (https://osduforum.org/). This is a data platform for sharing, communicating and doing analytics of data. It originated and has a foothold in the oil and gas sector but is moving into renewable energy and carbon capture installations too, as OSDU is a general framework for capturing any physical, geo-located asset (similar to Digital Twins).
The content can be adjusted based on the interests and background of a given student. But some interesting research questions to look into could be:
Some links: Lamina.ai with blog, LLMsPracticalGuide, and natural conversation.
The project can be linked to MIA Health if interesting for you.
A very recent paper supporting our initial thoughts and ideas behind the overall concept can be found here:
"Patients prefer ChatGPT over Physicians 79% of the time! A recent experiment used ChatGPT to answer medical queries on an online forum (Reddit). The responses were later graded by medical professionals and compared to responses from actual human physicians. As you can see in the charts below, the "robots" produced better-quality answers, and their answers also displayed more empathy."
Furthermore, the Med-PaLM 2 model just released scored 85.4% on the US Medical License Exam (USMLE) questions, i.e. expert knowledge (which is a 18% leap over previous state of the art results from Med-PaLM some six months ago, the first AI model that passed USMLE).
Interested? Get in contact so that we can discuss and design a project that would be really motivating for you to work on.
Life on earth displays an explosion of creativity and diversity. One single run of evolution has managed to create both photosynthesis, flight and human intelligence and is still presenting us with new solutions to the problem of survival and reproduction on earth. This capability for never-ending creation of novel organisms is what the field of open-ended evolution is trying to replicate.
One such open-ended evolutionary algorithm is the Paired Open-Ended Trailblazer (POET) algorithm. This is a coevolutionary algorithm seeking to endlessly generate problems of increasing difficulty and their increasingly complex solutions through the enforcement of a minimal criterion and goal-switching. POET has been applied to robot locomotion control and general game-playing environments.
A recent masters project has focussed on a new hyrbid algorithm involving POET and Neuroevolution of Augmented technologies (NEAT). An ablation study was performed to determine whether the algorithm’s minimal criterion and transfer mechanisms are in fact necessary, in thte light of a more challenging control problem with maze-like characterisitics with the introduction of adversaries.
This project is suitable for extensions, as is ie with such a hybrid algorithm as the basis, or to choose a different algorithmic approach to investigate open-ended evolution.
Artificial Intelligence is now being used at an increasing rate to augment or automate organizational decision-making. From processes such as performing credit checks on customers of banks, aiding in forecasting of future events, and automating manual and repetitive tasks, AI is introducing a new way of making decisions for organizations. The purpose of this project is to examine through empirical methods the effects and processes of transition to AI-based decision-making structures.
With the increasing use of video surveillance of public spaces, there is a growing need to take care of the privacy and integrity of those who are filmed by the surveillance cameras (Asghar et al. 2024). Full encryption of the video removes the ability to quickly find out what is happening on the video. The identity of persons who are not relevant to any investigation will then also be known by decryption. One solution to this problem is to partially encrypt the video.
In this proposal for a bachelor's project, we propose to encrypt parts of the video stream. There may be individual blocks in the video stream that contain faces or other things that can identify people. The main challenge in this project is to find a solution that does not require new players for the videos. The encrypted code must be inserted into the video stream in such a way that it is not corrupted at the same time that the hidden information cannot be extracted by unauthorized persons. To simplify the problem, one can choose to either encrypt, for example, license plates or faces.
The main objectives of this assignment are as follows
Particle Swarm Optimisation, where each particle may be thought of as an individual in a society of particles, provides a technique to model influences and effects in society that may prove beneficial in the battle against Climate change. This project is focussed on developing a PSO modelling environment that enables the study how an individual may become more inclined to increase their individual contribution/sacrifice for the benefit of future populations.
contact: Pauline Haddow
The task involves developing a domain specific conversational AI bot for a Hyper Interactive Intelligent Presenter. The bot would be trained to converse on a domain-specific topic for a course and will be useful to assist user in acquiring basic knowledge about a topic and/or for querying relevant material available on the course page. The bot is to be trained using deep learning on a variety of topics within a domain and be evaluated against different queries.
Current code generators, e.g., Copilot, generate code without considering a person's code styling. The code styles of the generated code could be very different from what developers used to do, and the typical optimization the developers used to put in the code could be missing.
This project aims to fine-tune a Generative AI-based code generator to generate code following personalized code styling and optimization of a developer based on the developer's historical code.
+ Literature review on approaches to identify code authors based on code styles.
+ Propose approaches to fine-tune Generative AI-based models to make it possible to generate personalized code, i.e., the generated codes have similar code styles as a particular developer used to have.
+ Evaluate the proposed approaches with open-source project data.
The human body constitutes a complex system wherein a large collection of parts interact. Consequently, medical interventions have varying effects on patients. Personalized Health is the field of research striving to predict strategies for medical interventions optimized for individual patients. In the scope of Personalized Health, recommender systems play a central role. They take a vast set of option and automatically narrow them down based on a user profile. The master project will pick a specific problem and implement a recommender system to personalize interventions. Candidate problems include, but are not limited to, nutrition, exercising, medication, and therapy.
The candidate should have attended courses about Artificial Intelligence, Machine Learning, and Recommender Systems. Ideally, the candidate has a vested interest in the topic. The candidate should have sufficient programming skills to process the underlying data, implement the algorithm, and develop a demonstrator. Experience with Recommender Systems and medical data is a plus.
Many patients need personalized training videos to perform rehabilitation at home. The current training videos from therapists are standardized, and do not fit individual needs.
This project aims to use sensor data and Generative AI technologies to generate personalized training videos based on the exercises the patients practiced during the clinical investigation.
The research questions are:
The project will be co-supervised by Prof. Frank Lindseth and Associated Prof. Gabriel Hanssen Kiss.
Several models can be used to find out how users’ social media networks, behaviour and language are related to their ethical practices and personalities, Such models include Schwartz’ values and ethics model and Goldberg's Big 5 model that defines personality traits such as openness, conscientiousness, extraversion, agreeableness and neuroticism. The thesis project would investigate applying such models to social media text and how the user personalties are reflected by the social networks that they participate in and develop.
Many regular maintenance operations occur over the lifetime of a commercial building. This includes for example replacement of air filters which filter the air supplied into a building. Short maintenance cycles stay on the safe side by replacing filters too often before any efficiency loss or down-time occurs. This may lead to time and material consuming replacements before they are actually necessary.
In an initial step, promising regular maintenance operations for automated prediction need to be identified and ranked based on their economic impact.
The goal of this thesis is to develop predictive maintenance methods for one or multiple of the identified operations in order to reliably detect the need for replacement or maintenance before a problem occurs.
This project is in collaboration with Piscada, a Trondheim-based technology company that develops an industrial cloud-based software platform for customers in construction and energy (PropTech), Industrial IoT, aquaculture, and general process management. The company was established in 2009 as a spin-off from SINTEF and focuses on innovation and simplification of industrial IT systems, as well as building a bridge between industrial automation and IT. There are today approximately 2,000 installations of Piscada's software and a diverse list of renown customers. We aim to be a leading industrial service platform with a focus on effective monitoring, new insights and optimization for increased sustainability in selected industries.
Many regular maintenance operations occur over the lifetime of a fish farm. This includes for example cleaning of the feeding mechanism or the tubes through which the feed is distributed to the fish-nets. Short maintenance cycles stay on the safe side by cleaning too often before any down-time or damage occurs. This may lead to time-consuming cleaning before it is actually necessary. Many fish-farm operators develop a good intuition for when a cleaning cycle is necessary, but this is not easily reproducible or transferable across employees.
In an initial step, promising regular maintenance operations for automated prediction need to be identified and ranked based on their economic impact. The goal of this thesis is to develop predictive maintenance methods for one or multiple of the identified operations in order to reliably detect the need for maintenance before a problem occurs.
This project is in colalboration with Piscada, a Trondheim-based technology company that develops an industrial cloud-based software platform for customers in construction and energy (PropTech), Industrial IoT, aquaculture, and general process management. The company was established in 2009 as a spin-off from SINTEF and focuses on innovation and simplification of industrial IT systems, as well as building a bridge between industrial automation and IT. There are today approximately 2,000 installations of Piscada's software and a diverse list of renown customers. We aim to be a leading industrial service platform with a focus on effective monitoring, new insights and optimization for increased sustainability in selected industries.
Anvendelsen av maskinlæring og AI for å predikere råvarepriser i markeder som opplever eller forventes å oppleve eksponentiell vekst.
The bioinformatics group works on developing and using computational models to predict how changes in gene regulation can control development and cause disease. Towards this end, we develop custom algorithms, statistical simulations, and machine learning-based solutions to analyze and interpret biological data; examples of previous MSc-theses include a genetic programming (GP) approach to predict microRNA target sites, a support vector machine (SVM) approach to identify microRNA genes, and an approach that combines GP and SVMs to identify related proteins.
We have several project and MSc-thesis opportunities for students interested in programming, machine learning, and string and search algorithms. If you are interested, please contact me to discuss specific projects.
DNV is currently leading a project under the auspices of ESA (European Space Agency) that focuses on the use of satellite data within shipping in the Arctic and Baltic Sea regions. The project aims to identify the needs for various types of satellite data, which services and products currently offering this, the extent and in which manner the satellite data is being used, and similar aspects. The current work on this project is published as reports on https://earsc-portal.eu/display/EO4BAS. The EO4BAS project is part of a larger project within EO data (Earth Observation, i.e., satellite data) financed by ESA and EC (European Commission). Not only opportunities within the maritime are explored, but also within ex. oil and gas, and raw material extraction.
One of the problems in publically available chest x-ray datsets is the lack of annotations/labels on x-rays for various abnormalities. For instance, the ISU chest x-ray dataset contains reports describing the issues present in the image, yet it is not easy to localize the areas having abnomalities on the images. Apart from the expert radiologists, it is even hard for the doctors to identify the abnormalities in x-rays. This project's objective is to use transfer learning approach to train deep learning models for similar abnormalities present in VinDR dataset or similar datasets to ISU for generating pseudolabels.
The gold standard for computer graphics is and always has been the simulation of real light dynamics through ray-tracing, but due to the high demands on compute it has seen little use in real time applications. In recent time ray traced real-time graphics has become a reality, with the last year even bringing ray-tracing capabilities to mobile devices. The advances that made this possible are several, including better process nodes for silicon, advances in neural network based denoising, novel temporal antialiasing techniques and improvements in Bounding Volume Hierarchy (BVH) construction. Ray tracing relies heavily on specialized data structures to make the intersection test between the ray and scene efficient using some kind of Acceleration Structure, with the standard approach being the use of a BVH. The BVH is a spatial data structure, and lends well to GPU warp-based execution because rays issued from nearby pixels (and scheduled on the same shader core) are likely to traverse the same part of the tree. This allows the tree nodes to be re-used for multiple threads of execution saving an order of magnitude in bandwidth. The difference between a well- and poorly constructed BVH can account for more than 50% of the ray traversal performance making quality a very sensitive topic. Another issue is that higher quality build algorithms naturally require more time, to the point where building the BVH takes too long to be feasible in a real-time environment. Due to this tradeoff between traversal performance and build time the field of acceleration structure construction is wide open and there are multiple heuristics that can be applied in attempt to get ahead in one way or another. The complexity of the problem is further increased by the fact that different hardware accelerators have different performance characteristics, meaning the same algorithm may not be the best everywhere. This all means that the construction of BVHes is not in general well understood, and there is ample room for innovation.
For the Fall project we would have you implement a standard Surface Area Heuristic (SAH) or Linear BVH (LBVH) algorithm and insert the resulting BVH in ARM’s hardware to evaluate performance. The initial implementation should be on CPU since it is much easier to program, but depending on how it goes you are encouraged to implement a build algorithm using Vulkan Compute as well. The goal is for you to have a solid base (and test pipeline) ready for your Masters thesis project, where you will have the opportunity to explore the various ways in which the standard algorithms can be improved upon. The specific direction here is up to the student. The algorithms that are implemented can be evaluated on their build time, their memory footprint and on their effect on the framerate of sample content. Several competing algorithms have implementations available on github and can be used for comparison.
Suitable for: 1 studentSupervisors: Theoharis Theoharis, NTNU, Torbjörn Nilsson, ARMRequirements: Computer Graphics courses (see below), Knowledge of C++, OpenGL, interest in learning VulkanCourses: TDT4195 (Visual Computing Fundamentals), TDT4230 (Graphics & Visualization), or equivalent.
Literature:1. https://jacco.ompf2.com/2022/04/13/how-to-build-a-bvh-part-1-basics/2. http://www-sop.inria.fr/members/Stefan.Popov/media/KDTConstructionGPU_TR10.pdf3. https://www.nvidia.in/docs/IO/77714/sbvh.pdf4. https://meistdan.github.io/publications/ploc/paper.pdf5. https://devblogs.nvidia.com/wp-content/uploads/2012/11/karras2012hpg_paper.pdf6. http://gamma.cs.unc.edu/SATO/SATO_files/sato_preprint.pdf7. https://vulkan-tutorial.com/8. https://research.nvidia.com/sites/default/files/publications/dnn_denoise_author.pdf9. http://behindthepixels.io/assets/files/TemporalAA.pdf
Digital transformation is influencing all the workplaces. Not always the digital transformation that is envisioned is successful, as witnessed by, for example, the challenges connected to the introduction of the Helseplatformen. One aspect that is often under-estimated is connected to the competences that are needed to workers to participate to the digital transformation in a meaningful way.
This task aims at designing a card-based co-design toolkit to help workers to understand the space of possibility of new technologies in their workplace. Focus will be on promoting creativity and system-thinking, at the same time keeping into account the constraints that are given by the existing infrastructures. The toolkit will take inspiration from Tiles (https://www.tilestoolkit.io/).
This task aims at designing a game to help workers to understand the space of possibility of new technologies in their workplace and their impact. Focus will be on systemic and critical thinking. Students are welcome to define, in cooperation with the supervisor, specific areas of interest, with respect to the learning objectives of the game, and game genre and technology.
Car tires have different features relevant for maintenance. These features include the profile depth, size, and manufacturer. Today, organizations tasked with tire maintenance need to keep track of these meta-data manually. In other words, employees have to inspect the tire and notes down existing damages as well as the meta-data.
The project's objective is to automate the process with the help of computer vision technology. The candidate will get access to a large collection of images of tires from different angles and meta-data that has been manually collected. Subsequently, the candidate will build computer vision models to predict the correct meta-data, estimate the profile depth, and detect possible damages.
The project is supported by Andreas Rosmo from Trønderdekk AS.
AI tools are widely spread and are used in different contexts, in schools, in the workplace, and as support to everyday practices. As a consequence, we all need to develop the digital competences that are necessary to use, understand, and influence the development of AI tools. This task aims at designing games for learning about AI and developing these competences.
Students are welcome to define, in cooperation with the supervisor, specific areas of interest, with respect to specific target groups (e.g. schools, specific workplace), the learning objectives of the game, and game genre and technology.
This topic is about creating a self-learning multi-agent scheme for steering simulated cars in an urban environment. The focus is on developing a Reinforcement Learning scheme for this application.
The project builds on existing results along the lines sketched above, and can be focused on different aspects, depending on the interests and the pre-knowledge of the student(s).
One possible focus is to perform the research in the well-known realistic autonomous driving simulator CARLA. Another possibility is to focus the investigation on using existing small floor robot platforms that we have, that is the LIMO robots by Agilex.ai
TL;DR: Approximate computing studies how to provide “good enough” results for a certain application. It is used in different context, for example resource constrained devices or when operating in degraded mode. We want to evaluate the impact of faults on different approximate computing techniques.
Approximate computing [1] is the science that studies how to provide “good enough” results -- according to an application-specific quality metric -- while at the same time improving a certain performance metric such as time-to-solution, energy-to-solution, etc. Many approximate computing techniques exist. In this project, we focus on compiler-level techniques.
Fault injection [2] is a verification technique that deliberately introduces faults into a system, to evaluate their effects. It is often used in the testing of critical systems, to create conditions to test redundancy and recovery mechanisms. Different techniques may be used, ranging from physical electrical disturbances to the alteration of source code. In this project we focus on injection at code level and during the compilation process.
While different approximate computing approaches may produce similar results in terms of quality and performance metrics, they may have very different behavior in terms of their ability to cope with faults (fault tolerance). The objective of this project is to apply fault injection to compare different approximate computing algorithms. This project will be based on existing approximate computing algorithms and existing benchmarks. The main task is to extend such benchmarks with a fault tolerance perspective, in which faults are injected to the existing code base.
The notion of responsible AI entails a large range of aspects regarding how AI applications are developed, utilized, and monitored throuhgout their lifecycle. The purpose of this project is to explore what responsbile AI means for organizations, which processes and structures they are establishing in order to attain set indicators of responsible AI, as well as what are the organizational impacts of it. Does adopting responsbiel AI result in any organizational gains? Does it influence how customers/citizens perceive the organization, or is it restricting what they can do with novel technologies?
Code obfuscators such as Binaryen are often used in security research to test detection capabilities of malware/cryptojacking code. A code obfuscator must tread the thin line between detection avoidance and efficiency (low-overhead).
In the first part of the project the candidate will conduct a literature survey and examine, among other topics, state-of-the art code obfuscation for Wasm.
The second part of the project aims to extend state-of-the-art security research into Wasm code obfuscation. A contribution could involve development of new security research tool(s) and toolchains, obfuscation methods or frameworks, or other proofs of concept (PoCs) with a specific focus on robust obfuscation at the instruction-level or intermediate representation. Binaryen or other LLVM-based obfuscation passes could be used as a starting point.
http://arxiv.org/abs/2401.05943
http://arxiv.org/abs/2403.15197
TL;DR: Evaluate different object detection and/or trajectory planing algorithm from the safey perspective. Need to know some machine learning for object detection. Builds on existing research and a previous Master’s thesis.
Autonomous vehicles rely on object detection as a fundamental way of perceiving the environment. Modular pipelines for autonomous vehicles first acquire data from sensors, perform object detection, create a scene representation, and finally perform motion planning.
Autonomous vehicles are safety-critical systems [1], where “safety” is defined as the absence of catastrophic consequences on the users and the environment. However, machine learning components are typically evaluated with traditional metrics based on precision and recall.
Recent works in the literature have proposed metrics and algorithms for object detection that integrate the concept of safety (e.g., [2]). The objective of this project is to evaluate the effectiveness of such algorithms on the trajectory planning step, from the safety perspective. The task involves a literature review on safety metrics for trajectory planning on autonomous vehicles. This work builds on existing research and on a previous Master’s thesis that has focused on the object detection step [3].
The long-term research objective linked to this activity is to define a benchmark that can assess the autonomous vehicles from a safety perspective.
Enhetlig pasientbehandling blant helsearbeidere i helsesektoren undergraves av ulike former for grenser - eografiske, institusjonelle og profesjonelle. Dette er til hinder for effektiv og høykvalitet pasientbehandling. Eksempler inkluderer samarbeide mellom fastlege og sykehus, eller samarbeide mellom sykehus og kommunehelsetjenesten herunder eldreomsorgen.
Trass pådriv og initiativ for å få helsearbeidere til å samarbeide tettere og mer interaktivt, gjenstår mye. Informasjonssystemene i helsesektoren er "silo"-orientert dvs de understøtter primært arbeidet lokalt, ikke samarbeide gjennom behandlingskjeden.
Det har opp gjennom årene vært satt i gang en rekke reformer og tiltak (feks Samhandlingsreformen, En innbygger en journal, Helseplattformen) uten at dette har løst utfordringene.
IKT (digitalisering) blir pekt ut som mulig løsning, gitt kapasitet til støtte distribuerte arbeidsprosesser.
Prosjektet/ oppgaven vil ta for seg et utvalgt innføringsløp for en digital tjeneste i helsesektoren. Oppgaven vil innebære en selvstending, empirisk innhenting av krav gjennom observasjon, intervju og logging av bruk av eksisterende system. Krav/ behov skal så operasjonaliseres i anbefalte, ev også prototypet, funksjonalitet.
Football has become an enormous industry where players are bought and sold based on information about them. This information is not just collected during games, but during trainings, tests, leisure and even when sleeping. In the wrong hands, this information can be used for extortion, manipulate the market or influence betting odds.
In 2022, the player representative organisation FIFPRO launched the the Charter of Player Data Rights. This charter aims to protect the privacy of professional footballers and allow them to benefit from personal rights to manage and access information about their performance and health. However, there is a lack of available implementations for the protection, collection and use of player data. For instance, players do not appear to have sufficient means to control the collection of their data and they are also concerned with the portability of moving their data from one club to another. Also, such solutions should become available and affordable not only to top players, but amateurs and youths as well.
This project will investigate how technical solutions can support player data rights, and at the same time support positive usage of this data (improve player development, prevent injuries, engage fans, economic compensation). The student(s) will do a literature study and develop threat models to potential solutions. In a master thesis continuation, a proof-of-concept should be developed and tried out with representative end users.
Note: The specific details of the assignment can be tailored according to the student's interests and expertise.
Home Automation Systems are commonly found in many homes and are there to make our lives easier, but like any other connected devices, are vulnerable to various security threats. Some of the common security challenges associated with home automation systems are:
The goal is to analyse the software security of the popular open source project Home Assistant (https://www.home-assistant.io/), in particular the integration with the emerging Matter standard (https://csa-iot.org/all-solutions/matter/). The student(s) will suggest security improvements and possibly contribute to the open source project.
Human activity recognition refers to the task of automatically identifying and classifying different activities that a person is performing, such as walking, running, sitting, or standing, based on data collected from various sensors.
The Transformer model, originally introduced for natural language processing tasks, has shown remarkable performance in capturing long-range dependencies and patterns in sequential data. This project focuses on developing the Transformer architecture to process and understand sensor-generated sequences, which could include data from accelerometers, gyroscopes, and other wearable or environmental sensors.
The project will be carried out in collaboration with Benjamin Cretois at the Norwegian Institute for Nature Research (NINA). NINA is Norway’s leading institution for applied ecological research, with broad-based expertise on the genetic, population, species, ecosystem and landscape level, in terrestrial, freshwater and coastal marine environments. The student(s) will be collaborating with researchers at the Miljødata department who focus on employing advanced technologies to study and protect biodiversity, with a particular emphasis on bioacoustics and bird species.
NINA are currently using an off-the-shelf deep learning-based bird classifier (BirdNET) that effectively identifies common bird species. However, its performance significantly drops for the rarer species, limiting our understanding of biodiversity in diverse ecosystems. Properly accounting for rare species is critical to establish ecosystem health and it is essential to develop automatic methods efficient at detecting and classifying them.
The project aims to refine the model using advanced machine learning techniques such as fine-tuning and few-shot learning strategies, potentially exploring other innovative methods to enhance species recognition, particularly for underrepresented species. The expected outcome includes improved model accuracy and recall for rare species, contributing to more comprehensive biodiversity assessments.
The student(s) will have access to the Sound of Norway (https://thesoundofnorway.com) dataset. It contains approximatively 7 terabyte of audio files recorded by BUGG recording devices at 41 different sites throughout Norway. All of the detections made by BirdNET have been verified by an ornithologist. Moreover, a few files have completely been reviewed by the ornithologist to compute the model recall. This dataset should provide a robust foundation for evaluation of the methods that will be developed.
Risks / Challenges:Improving model capabilities to generalise well to rare events (rare bird calls in the case of this subject) is a hot topic and numerous methods are being developed. The primary scientific challenge lies in the method's ability to generalise well to rare species with limited examples.
The project is aimed at the automatic classification of sentiment in texts on Twitter or other social media, tentatively addressing issues involving negation and/or figurative language, that is, language which intentionally conveys secondary or extended meanings (such as sarcasm, irony and metaphor). Figurative language creates a significant challenge for sentiment analysis systems, as direct approaches based on words and their lexical semantics often are inadequate in the face of indirect meanings. A subgoal of the project would then be to find a set of tweets rich in figurative language, but the main goal would be to determine whether the writer of such tweets has expressed a positive or negative sentiment (commonly displaying positive or negative emotions towards a product, person, political party, etc.), and possibly the degree to which this sentiment has been communicated.
More than 80% of our global goods are transported by ships. Like the goods they transport, ships will eventually become waste and need to be broken down properly. Ship breaking is a complex and dangerous job since ships contain different types of toxic materials that should be properly located, identified, and removed to prevent adverse effects on humans and the environment.
Currently the planning work before breaking down ships is done using 2D images printed on papers, which show the ship layout from different angles and hazardous materials that exist inside. However, this practice is not entirely efficient and safe because, once the cutting has started, there is no way to reverse the action in case there is a mistake or safety issue. Therefore, there is a need to develop a shipyard simulator that visualizes 3D models of ships and supports user inputs, such as defining where the ship should be cut and visualizing the outcome if the defined cutting plan is approved.
This project is about the inspection of a ship tank or similar "environment" using a drone, and aiming providing contributions for making this inspection process fully autonomous -- a task which of course cannot be achieved with a single student project, thus: "contributions to...". In this project we assume that we have some 3D model of a industrial tank, and a set of waypoint coordinates that an inspection drone should visit. The goal of this project is then to train a model to perform short-term path planning between these waypoints using reinforcement learning. In the current concept, a hybrid approach fusing both (deep) reinforcement learning as well as model-based predictive control (MPC) is considered; the design decisions are however made after an in-depth literature study has been performed. The resulting planning module should propose collision-free efficient paths for the drone to execute, and these short-term plans are to be evaluated in simulations, using them in a receding horizon predictive control scheme.
Languages change rapidly over time and language users adapt to different situations and setting. Studying how language and communicative processes evolves is a highly multi-disciplinary task involving machine and language learning as well as biological and cultural evolution. The aim of this project will be to use methods such as evolutionary algorithms or reinforcement learning to investigate the main dynamics in language evolution.
When trying to understand the origins of languages, we can compensate for the lack of empirical evidence by utilising evolutionary computational methods to create simulations of how language may have evolved over time, e.g., by creating "language games" to simulate communication between agents in a social setting. In general, simulations on language evolution tend to have relatively small and fixed population sizes, something this study could aim to change.
The project aims to study various aspects of creating a solution that facilitates sharing office/desk use, converting them into “smart” desks or “context-aware” desks.
The project involves a study of relevant existing research and literature, designing, implementing, and evaluating prototypes ( IoT + software), and planning and conducting a series of user tests.
While the project can be assigned to a single student, it is recommended that a pair of students will work on it.
Smidige metoder oppstod i team som jobbet med systemutvikling, men har blitt tatt i bruk av større deler av organisasjoner og ført til endringer i ledelse av prosjekter og større programmer. Smidige metoder legger vekt på beslutningstaking på lavt nivå i selvstyrte team, tett kontakt med kunde og fleksibilitet i arbeidsprosesser for å kunne håndtere endringer underveis i utviklingsprosjekt. Når organisasjoner ønsker å ta i bruk denne type metoder i større deler av organisasjonen støter de ofte på utfordringer med eksisterende hierarki, at det kan være vanskelig å bli enige om arbeidsprosesser med miljø med andre oppgaver enn programvareutvikling og generalt at det er motstand mot organisatoriske endringer. Denne oppgaven vil bestå i å først gjøre et litteraturstudie på temaet “smidig transformasjon” innen fagområdene systemutvikling og informasjonssystemer, og koble med relevante andre felt som for eksempel endringsledelse. En videreføring i masteroppgave vil kunne innebære et empirisk studie av en eller flere organisasjoner som gjennomfører en “smidig transformasjon”.
For en innføring i forskningsfeltet, se forskningsagenda publisert på workshop på den internasjonale konferansen om smidig metodikk: https://link.springer.com/chapter/10.1007%2F978-3-030-30126-2_1
En masteroppgave kan delvis basere analyse på funn fra kompetanseprosjektet Agile 2.0, hvor NTNU var en partner sammen med DNV GL, Equinor, Kantega, Kongsberg Defence & Aerospace, Sopra Steria, Sticos. Prosjektet ledes av SINTEF Digital og er støttet av Norges forskningsråd.
Robots have become part of our daily lives. Some work in factories while others vacuum our homes. The research field of Social Robotics studies how people interact with robots. Specifically, we want to advance robots to complement us and facilitate communication. The NorwAI Center has access to robots that have sensors for audio-visual input. We are excited to explore new possibilities to address real-life needs.
The candidate(s) will have the chance to use the robot(s) to develop new functionality targeting Human-Robot interactions. We want the robots to be able to communicate in Norwegian and English. Student researchers have already created a modular backend that helps to integrate new functions. We have a set of ideas, but the candidate(s) will have the opportunity to contribute, implement, and evaluate ideas themselves.
The task demands solid programming skills and a keen interest in HCI, particularly natural language processing. The robots are occasionally showcased at events, which can be used for evaluation.
Cryptic crosswords are puzzles that rely not only on general kwowledge, but also on the solver’s ability to manipulate language on different levels according to Sadallah et al., 2024 https://arxiv.org/pdf/2403. 12094. This is a hard problem, which requires much reasoning on the part of humans. The authors report an accuracy of 8.9% for the best Large Lan- guage Model (LLM), wheras human performance is 99%.
As a consequence, this application area is a great opportunity to explore the potential of LLMs in solving difficult reasoning tasks.
In this project, candidates will be refining the methods employed above, by benchmarking the LLMs produced at NorwAI agains the state-of-the-art models used in the above paper. Datasets for English also exist, but producing datasets for Norwegian is a second goal of the project. This can be done synthetically, and also in cooperation with rights’ holders.
Neural attention methods have been the new motor for artificial intelligence. However, the most popular attention model, Transformer or its variants, suffers from the quadratic computational cost and is thus expensive in practice.
In the project, we will study a new neural network for self-attention that has a much lower computational cost and better transformation capability. The developed model will be tested on data from various domains.
Strong programming and mathematics skills are essential for this project.
The goal of this project is to explore the application of models commonly used for predicting species discovery to the task of identifying vulnerabilities in software systems. Drawing parallels between the process of species discovery and software vulnerability detection, the proposal is to develop or adapt models inspired by species accumulation curves to analyse the cumulative number of software vulnerabilities discovered over time. By utilising historical vulnerability data and considering factors such as software complexity and codebase size (if available), these models will seek to predict the rate of new vulnerability discoveries and estimate the total number of vulnerabilities within a software system. Experience with statistical models and methods will be instrumental.
The research outcome would aid software vendors and security researchers estimate the number of vulnerabilities in software based on historical data, as well as adapt new strategies for bug hunting when current methods predict a low number of future discoveries.
https://www.youtube.com/watch?v=TeY1fY0Bi_M
https://scholar.google.co.uk/citations?view_op=view_citation&hl=de&user=bX-GbkUAAAAJ&citation_for_view=bX-GbkUAAAAJ:hFOr9nPyWt4C
One of the key chellenge we aim to address in this project is to predict the shot power of a bat hitting the ball (baseball /cricket) along with the type of the drive/stroke, and bat angle from the video clips.
We have created a labelled database containing drive/stroke classes that we want to enrich with additional analytics using computer vision techniques and deep learning algorithms via Vision Transformers. The selected student is required to develop and apply various models estimating shot power, type, and bat angle from recorded clips.
In AI and ML, several methods rely on stochasticity or randomization: mutation and crossover in evolutionary algorithms; dropout and stochastic gradient descent in deep learning; stochasticity in stochastic local search (SLS); and randomization in systematic search. SLS algorithms, which we study here, are competitive in solving computationally hard problems such as satisfiability (SAT), sparse signal recovery, scheduling, and most probable explanations in Bayesian networks (BNs). Essentially, SLS algorithms are greedy optimizers that also make random moves in order to avoid getting trapped in local but non-global optima. Further, SLS algorithms are interesting in that they can studied formally, for example by means of Markov chains.
The goal of this project and thesis is to study the theory and application of SLS. The following research questions are examples - specifics depend on finding a topic of broad and common interest. First, there are opportunities to improve SLS, with an eye to specific applications. Can this done by combining different heuristics and adaptive methods, perhaps including concepts from bio-inspired AI including evolutionary algorithms?
Second, the computational cost of ML can currently be high, and the use of SLS for ML, for example for feature selection by means of wrapper methods, is no exception. Maybe Markov chain-based analysis and/or high-performance computing can be the basis for SLS methods that better handle massive and complex datasets while reducing computational cost dramatically?
Third, the theory and formal methods communities have also studied parametric Markov chains, but interactions with the AI and ML communities have been limited. Perhaps this will change, now that trustworthy AI is coming to the forefront?
Immersive technologies such as virtual/augmented/extended reality (VR/AR/XR) have demonstrated significant potential in transforming educational practices by providing learners with realistic and highly engaging learning experiences. In most cases, due to budget and practical concerns, educators use relatively unexpensive XR equipment such as Oculus Quest. While this might be sufficient for many educational situations, it is important to investigate the potentials of more advanced equipment that provides advanced spatial computing possibilities, simulates senses other that sight and hearing and facilitates walking.
The goal of this project is to explore how advanced XR technology, beyond the regular XR equipment, could support learning, especially at university and professional level. The specific topic of the project will be defined in collaboration with the student depending on the choice of equipment. Here are examples of possible projects:
One of the core competence areas of professional involved in the development of IT is the identification and representation of requirements. For modern IT-solutions sustainability can be considered a key concern, thus we have to look upon how to achieve sustainability by design, and not as an afterthought. This project will investigate areas related to
• How to express sustainability requirements to software and information systems• How to elicit sustainability requirements to software and information systems• How to balance sustainability requirements and more traditional form of requirements• How to deal with sustainability requirement in modern agile development methods• How to evaluate adherence and achievement to sustainability requirements on a short and long time period
Typically the task will start with a structured literature review in the autumn, with a follow up in the spring evaluating the use of existing or new techniques for handling sustainability requirements in agile software development. The task will be related to NTNU Center for Sustainable ICT (CESICT), and the national network goforIT (Grønn omstilling for IT-bransjen)
Cloud computing is increasingly popular due to its potential benefits in scalability, cost savings, and flexibility. However, it also poses the risk of vendor lock-in, where an organisation becomes dependent on a specific cloud provider's proprietary technology. This dependence can again limit flexibility, increase costs, and hinder migration of software defined infrastructure to more sustainable cloud solutions [1]. To enhance the robustness of their systems, businesses are also looking for multi-cloud supported solutions.
To address these challenges and ensure the sustainability of cloud infrastructure, this master's project will investigate methods and strategies to mitigate the risk of vendor lock-in. The project will review the current state-of-the-art in sustainable cloud migration and explore practical solutions to reduce vendor dependency.
[1] J. Opara-Martins, R. Sahandi and F. Tian, "Critical review of vendor lock-in and its impact on adoption of cloud computing". i-Society 2014. doi: 10.1109/i-Society.2014.7009018.
Company intro
SpareBank1 SMN is the leading financial institution in middle Norway. SpareBank 1 SMN is an independent regional savings bank with a local footing and one of six owners of SpareBank 1-alliansen. SpareBank 1 SMN offers competitive products in the fields of financing, savings and investment, insurance and payment services along with estate agency, leasing, accounting and capital market services. The thesis will be carried out in the advanced analytics department in the technology and development division. Relevant business resources will be included if necessary.
Contact person at SMN1: Stian Arntsen, Stian.arntsen@smn.no
Banking data can contain sensitive information that limits the possibility of sharing the data openly. The data is typically restrained to being analyzed on our internal hardware. This limits the ability to use large compute clusters and cloud services. It also sets a high barrier of entry for using the data in research than what an open dataset would. If synthetic data could be created based on real data, which retains the inference and statistical relationships without containing personal data, this could make it both easier to share data and to use compute resources outside of our servers.
Methods for generating synthetic data include generating data from statistical distributions, using agent based modelling, GANs and LLMs. Text data that contains personally identifiable information is hard to synthesize. Transactional data, such as card transactions can be easier to remove any personal information from, but it will probably be hard to maintain statistical relationships, especially in tandem with other customer data. One approach could be to try to synthesize an already aggregated dataset, that would contain for example the sum of transactions for a given category such as restaurants. A younger person might have a very different history of transactions. In that case, a GAN should be able to create datasets that retain the relationship between age and restaurant expenses. Another approach could be to synthesize different raw transactional data, that is connected to customer data and product data in a relational form. In that case, it would be more challenging to generate data where these statistical relationships are retained, but this would also open up a lot more use cases than synthesizing a single table or flat file. We expect the student to synthesize banking data in such a way that the data will retain statistical information while not containing any real data. The synthesized dataset should give the same statistical inferences as the original data
The data would either be an already aggregated flat file, for example one of our analytical base tables for next best offer models, or a subset of relational data, such a transaction data, product data and customer data. The latter would be more challenging but also open up a lot more use cases. Students working on our data will be working on a virtual machine through a company laptop. All the data processing will have to be done on the virtual machine. The data will be de-identified.
Forskning ved psykologisk institutt indikerer at en underliggende årsak for blant annet dysleksi og dyskalkuli kan være dysfunksjon i visuell prosessering. Nærmere bestemt er det nerveceller som flytter informasjon saktere hos disse gruppene enn hos de som ikke har disse utfordringene. Disse nervecellene kalles magnoceller og har i hovedsak ansvar for å oppfatte raske forandringer i omgivelsene.
Et system som kan bruke til å teste visuell informasjonsbearbeiding for å finne tegn på slike forstyrrelser er utarbeidet, og oppgaven går på å evaluerer og videreutvikle denne slik at det kan brukes i praksis for testing av barn.
Hovedideen er beskrevet her: https://gemini.no/2020/01/synet-den-egentlige-arsaken-til-barns-problemer/ (denne lenker også til en artikkel som kom i Scandinavian Journal of Educational Research) i 2020, og gjøres i samarbeid med Hermundur Sigmundsson ved psykologisk institutt.
Vurdering og tilbakemeldinger er viktig for læring og i prosjekt og programmingsfag er det behov for mange former for tilbakemeldinger enten fra fagstab, fra medstudenter, eller basert på automatisert testing og analyse. I dette prosjektet skal studenten(e) jobbe med forskjellige elementer av et innleverings- og tilbakemeldingssystem.
Funksjonalitet som kan være i fokus for oppgaven er:
In Norway we have a well-developed standard for naming equipment and components in buildings, TFM. However, abroad there is no such standard and many different conventions are created and used.
In this project the student will use our properly labeled (ground truth) time series data to build a model that can classify equipment based on the data they've emitted the past two weeks or less. The student will have to settle on a suitable way to preprocess the time series and type of model to use. Also, perhaps evaluate if the classes should be joined or divided based on clustering. This is challenging because different buildings may have different patterns of operation and setpoints.
The student will either get access to our API to fetch data or a hard drive containing the data. He/she will also have a list of labeled tags (sensors). We have years of high frequency data from hundreds of
buildings. The data quality in Building Automation Systems is generally very good. The data will be anonymized but not require any further level of protection.
Teamwork is central in software development, and is currently a topic much addressed in agile software development where the development is performed in small, self-organized teams. Improving the efficiency and effectiveness in software development will therefore often involve improving the way the teamwork is organized. Several team performance models have been suggested in the research literature, and there is a growing number of empirical studies of teamwork in software development with focus on specific characteristics for agile development teams and distributed or virtual teams.
This assignment will involve making a literature review of research on teamwork in software development, and can be continued in a master thesis with an empirical study of factors that effect team effectiveness in software development teams. For an overview of previous findings on teamwork, see: ieeexplore.ieee.org/abstract/document/7498535
A large variety of algorithms have been published to date which allow for 3D surfaces to be compared to one another. These are primarily used to detect point to point correspondences between two potentially similar surfaces. Detecting similarity is the foundation for many things you'd want to do with 3D data, such as an autonomous robot being able to recognise places it has been before.
One problem with all of these methods is that they usually require a scale parameter to be defined. The scale of recognising a chair would for example be different than that of a building. Existing work sidesteps this problem by manually setting this parameter. However, it stands to reason that the optimal value of this parameter could be estimated automatically in some way.
The objective of this project is to develop an algorithm that allows the scale parameter of these recognition methods to be selected based on the surfaces being matched. The largest problem to solve is that for two corresponding points on two similar surfaces the estimated scale needs to be the same, even when one of the two surfaces is altered in some way.
While there has been a long discussion about the potential of using Artificial Intelligence in private organizations, now more and more public organizations are implementing solutions to support their operations. From uses for fraud detection, chatbots, autonomous vehicles, or infrastructure monitoring, AI is gaining ground in applications for public administration. This project will be done in connection with SINTEF Digital and will involve data collection, analysis and reporting. The aim is to find out what is the status of AI adoption, what are the potential interesting uses, and what is the value that is realized.
Today more and more companies are using big data analytics to support or drive their business strategies. Yet, there is ongoing debate about whether such investments do indeed create value if so how this value can be captured. The objective of this master thesis is to perform a quantitative study on companies in Norway and examine the ways in which they are applying big data analytics to create business value. The project is in cooperation with the Big Data Observatory (https://www.observatory.no), during which you will learn how to develop research methods and analyze quantitative data.
The primary objective of this project is to use large sets of health data to generate prospective futures/forecasts for individual patients. These forecasts, while not strictly evidence-based or derived from clinical best practice, outline potential future health-related events, diagnoses, interventions and contacts. Such as a data-based forecast that facilitates discussing potential choices and promoting shared clinical decision-making. A particular use case is to combine these forecasts with actual evidence-based recommendations in order to aid school pedagogical-psychiatric services and general practitioners in taking action or preparing referrals to specialist CAMHS clinics. Specifically, we aim to provide ranked forecasts and evidence-based recommendations according to patient relevance and other criteria.
Approach
To achieve the above objective, students are free to use any appropriate approach. However, the work would be hands-on with real health data, including data preprocessing, mining, model training, recommendation generation, and validation and verification with clinician involvement.
The student will be supervised by:Main supervisor: Øystein Nytrø (https://www.ntnu.no/ansatte/nytroe)Co-supervisor: Dipendra Pant (https://www.ntnu.no/ansatte/dipendra.pant)
Supervisors: Sofia Papavlasopoulou, Isabella Possaghi, Boban Vesin
Suitable for: One or Two Students
This thesis focuses on designing and implementing a digital toolkit that guides and collects data on technologically rich Design Thinking activities (Empathize, Ideate, Define, Prototype, Test) for K-12 students. These activities focus on collaborative problem-solving approaches to generate inventive solutions to real-world issues with the help of digital tools. Specifically tailored for K-12 students, the toolkit will serve as a means to assist students in their Design Thinking tasks connected to learning and gather insights into their performance. The aim is to create a toolkit that can support students in brainstorming and ideating powerful concepts and in their following prototyping with technologies and testing, all while gathering information on their creative processes. The focus is to support the student’s learning process and the collaborative dimension among peers.
Thesis DescriptionThe first stage of the candidate(s)' work involves conducting a comprehensive review of current literature. This process aims to gain insights into the discipline of Design Thinking (and other project-based approaches), its current application in educational settings, and existent toolkit functionalities that support the learning processes. Following this step, the candidate(s) will shape their contribution based on the identified novelty, best practices, and specific requirements for the toolkit that can bring innovation. Along with the design of the toolkit (in the form of a platform or a website) performed with an iterative approach, the candidate(s) will focus on data gathering, storage and analysis. As a following step, the candidate(s) will perform a testing session with target users, in an on-site setting, to assess the efficacy of the proposed toolkit and collect data. Collected user data and its analysis will assess the toolkit's validity. The toolkit will be refined based on the findings. Finally, the candidate(s) will analyse the data to complete the design journey and lay the groundwork for writing the thesis.RequirementsThe ideal candidate should have a background in software and interface design, as well as good programming skills, particularly in JavaScript. Familiarity with HTML and CSS would be nice to have. Moreover, an interest in user-centred research and hands-on methodologies is sought after.
Description in which company/unit the thesis will be placed: SINTEF Digital, DNV, Kongsberg, NorwAI: In hybrid AI the aim is to combine machine learning and physical knowledge. This is particularly relevant for industrial applications of AI and machine learning on physical systems. Through the SFI NorwAI we work on hybrid AI together with partners in the energy sector. We have two use cases focused on predictive maintenance of wind turbines and on virtual flow meters in oil and gas. Problem Description: Industrial data are typically consisting of time series, 3D models, and documents where relevant features have to be identified and extracted automatically. While deep learning approaches have been successfully used to this extent, geometrical and topological methods offer a compelling alternative. These are typically robust to noise and offer a more apparent interpretation and explanation. Thesis Description: At first, the performance and sensitivity of topological data analysis shall be tested on industrial data. Approaches based on algebraic topology such as persistent homology shall be tested on time series data for the detection of specific patterns, associated with industrial processes or equipment faults (condition-based monitoring) for machine learning classification. TDA could be also benchmarked against deep learning methods. Another possibility: similar topological approaches based on homology can be applied to more complex geometric data types such as CAD 3D models of industrial equipment (piping, pumps, tanks, etc.), in order to automatically identify the geometry of the equipment. As this task is computationally demanding, different sampling and averaging approaches based on Radon-measures could be tested [https://arxiv.org/abs/2212.08295] or grid sampling based on eigenfunctions of the Laplace-Beltrami operator. The project is open-ended and flexible to accommodate the ideas and interests of the students. The project can be adapted to students from several departments such as Mathematical Sciences, Computer Science or Engineering Cybernetics Data Description: The project is based on a mix of public benchmark data and proprietary simulations from industrial partners (from Cognite), as well as operational data. This will require approval from the partners before publication.
Industrial assets’ condition monitoring holds immense potential for reducing downtime and operation costs. However, this task is very complicated and requires domain expertise in addition to advanced data analysis tools. While deep learning has been successful in several condition monitoring applications, this lacks explainability and requires a large amount of labelled training data on the many possible failure modes of each piece of equipment. As well, limited benchmarking datasets have hindered the development and validation of effective condition-monitoring systems.
This thesis will focus on Topological Data Analysis (TDA) and other methods based on algebraic topology. This offers an intriguing alternative to traditional Fourier methods in the realm of equipment condition monitoring. Unlike Fourier methods, which primarily focus on frequency domain analysis, TDA enables the exploration of complex geometric structures within the “shape” of data sets, providing a unique lens through which to understand equipment failure patterns. By representing equipment failure data as a topological space, TDA allows for the detection of subtle, non-linear relationships and structural changes that may go unnoticed in traditional analyses. This approach has the potential to uncover hidden patterns, improve predictive models, and enhance the accuracy of failure predictions, making it a compelling and promising avenue for advancing the field of equipment condition monitoring.The candidate will:
I. Apply methods based on persistent homology, cohomology and other recently published methods on industrial data to try to identify failures early. Candidate systems could be toy models, gearboxes, ball bearings in wind turbines, motor failures in pumps, slugging and other flow assurance issues for offshore O&G installations and pipelines. II. Have the opportunity to develop data pipelines and procedures for predicting failure of heavy industry equipment, possibly comparing them to established deep learning methods. III. Alternatively propose and test topology-based methods themselves, depending on their interest and background. We are open to suggestions.
Cognite will provide data sets from use cases such as Wind Turbine monitoring, Ball Bearings, or Oil & Gas events.
Simone Casolo (Cognite) will support this master project.
The candidate ought to have good knowledge of python programming. The underlying mathematics are not trivial.
Traditionally, software development and operations have been handled by separate teams. In such an organisation, a central objective of the development team is to introduce new application functionality, something that conflicts with the operation team's goal of ensuring stable and reliable services. The conflict in terms of the functions and values that the two teams provide often lead to the implantation of rigid processes that delay development. Hence, recent approaches aim to break down these barriers, forming multidisciplinary teams (DevOps) that are responsible for the entire lifecycle of an application, including operations and maintenance. Organising people from both development and operations in such teams, while also automating parts of the deployment process, has showed to both enable rapid deployments of new functionality and to increase software quality [1].
However, many organisations still struggle to adopt DevOps practices [2] and transforming traditional software and IT operations into DevOps teams is not straightforward. This project will investigate how organisations have managed the transition to DevOps, and how the different functions traditionally provided by an IT operations team can be re-organised in a DevOps development model.
[1] Forsgren, N. et al. Accelerate: The Science of Lean Software and DevOps: Building and Scaling High Performing Technology Organizations. 2018. [2] Krey, M. et al. DevOps adoption: challenges & barriers. HICSS2022. https://digitalcollection.zhaw.ch/bitstream/11475/24225/3/2022_Krey-etal_DevOps-Adoption_HICCS.pdf
Problem description TrønderEnergi has several forecasting systems utilizing machine learning methods in commercial operation. The systems are used and the forecasting is monitored by domain experts in the operating center at Berkåk. The domain experts do not have any training and knowledge in the machine learning methods, but they do have to make decision based on their output. At times the experts need to understand why a forecast has the value it has. The literature has to be reviewed, prototypes should be developed and evaluated by domain experts.
Data There are several data sets that can be used for this including consumption, grid loss, weather forecasts, and power production data. Many of the data sets have more than five years of data.
The focus will be on data from deployed systems, so that they can be improved.
Challenges Provide explanations to domain experts, so that they understand the forecasts.
This project aims to address the challenge of Uncertainty Quantification (UQ) for Language Models [1], recognizing its crucial role in ensuring confidence and reliability in the generated text.
The increasing adoption of Language Models in safety-critical applications highlights the pressing need to assess and quantify the uncertainties inherent in their outputs. While Language Models have demonstrated remarkable capabilities in natural language generation, their outputs often lack transparency, raising concerns about their reliability in critical scenarios. Uncertainties in Language Models can stem from various sources, including data limitations, model architecture, and the inherent stochastic nature of language generation.
Ensuring the reliability of Language Models in safety-critical systems requires effective methods for uncertainty quantification. However, achieving robust UQ for Language Models presents significant challenges. Existing approaches often lack scalability, struggle to capture nuanced uncertainties, or require extensive computational resources. Additionally, the interdisciplinary nature of the problem necessitates a thorough understanding of both language modeling techniques and uncertainty quantification methodologies.
The student is expected to review relevant literature on UQ for Language Models to gain insights into the strengths and limitations of current UQ techniques and identify promising avenues for further research. The student will implement and test selected Language Models (e.g., deep learning [2], Tsetlin machines [3], probabilistic models [4]) and UQ methods on benchmark datasets commonly used in natural language processing tasks. This empirical evaluation will provide insights into the effectiveness of different approaches in capturing and quantifying uncertainty in Language Models generated text. Furthermore, the student is expected to critically reflect on how UQ can address challenges associated with Language Models, such as hallucination, misinformation, and lack of interpretability. By examining the role of UQ in mitigating these issues, the student will contribute to advancing the understanding and application of UQ techniques in the context of Language Models.
To investigate uncertainty quantification for Language Models, the project will leverage publicly available benchmark datasets [5] commonly used in natural language processing research. These datasets will provide a standardized framework for evaluating different UQ approaches across various tasks, including text generation and summarization. The selected datasets will encompass diverse linguistic phenomena and application domains, enabling comprehensive assessment of uncertainty quantification methods.
DNV is a risk management and assurance company. This means that we help our customers across various industries to make sure that their products and processes are safe, sustainable, and compliant with regulations. The Digital Assurance program in our R&D department is tasked to explore new digital technologies that DNV can use to provide better assurance or may need to assure in the future. This includes the use and assurance of AI-based systems.
[1] Chang, Y., Wang, X., Wang, J., Wu, Y., Yang, L., Zhu, K., ... & Xie, X. (2023). A survey on evaluation of large language models. ACM Transactions on Intelligent Systems and Technology.
[2] Abdar, M., Pourpanah, F., Hussain, S., Rezazadegan, D., Liu, L., Ghavamzadeh, M., ... & Nahavandi, S. (2021). A review of uncertainty quantification in deep learning: Techniques, applications and challenges. Information fusion, 76, 243-297.
[3] Abeyrathna, K. D., Hafver, A., & Edward, L. Y. (2023, August). Modeling prediction uncertainty in regression using the regression tsetlin machine. In 2023 International Symposium on the Tsetlin Machine (ISTM) (pp. 1-8). IEEE.
[4] Hwang, J., Gudumotu, C., & Ahmadnia, B. (2023, September). Uncertainty Quantification of Text Classification in a Multi-Label Setting for Risk-Sensitive Systems. In Proceedings of the 14th International Conference on Recent Advances in Natural Language Processing (pp. 541-547).
[5] Lhoest, Q., del Moral, A. V., Jernite, Y., Thakur, A., von Platen, P., Patil, S., ... & Wolf, T. (2021). Datasets: A community library for natural language processing. arXiv preprint arXiv:2109.02846.
The ability to account for uncertainty becomes important when applying AI in domains where wrong predictions can have severe consequences. Uncertainty-awareness enables the AI (or those who use it to support decisions) to apply caution when uncertainty is high in order to reduce risk. The goal is to avoid that an AI is “confidently wrong”. Many approaches exist to make AIs aware of uncertainty (e.g. Bayesian deep learning, Gaussian Process models, Monte Carlo dropout and Evidential deep learning). However, how can we know that the uncertainty expressed by such models is meaningful? If the perceived uncertainty is misguided, then an AI may still be confidently wrong or overly conservative. One step towards answering this is to investigate what determines the uncertainty of the AI.
Standard deep learning (DL) classifiers use a softmax layer for the output, which means that the probabilities for the training classes always add to 1, even if an input depicts neither of the training classes. This can lead to the classifier being confident that an image belongs to a certain class even if it shows something completely different. The student should review methods that exist for making DL classifiers uncertainty-aware and able to tell when a sample does not belong to any of the training classes. The student should implement some of these methods and test the performance when the classifiers are given images from classes not contained in the training data or manipulated images from the classes contained in the training data. The student should investigate the type and extent of manipulation (blur, contrast changes, color shifts, rotations, translations, overlays, etc.) that is necessary to confuse the classifiers and their ability to tell when a picture belongs to an unseen class or is too distorted to be classified.
Data Description:
DNV will make available an in-house dataset containing about 3000 labeled images of various types of boats. In addition, to enable comparison with other research, it is suggested that the student test out models on other openly available datasets such as RETINA benchmark and the CIFAR-10 and CIFAR-100 datasets (e.g. comparing out-of-class misclassifications between CIFAR-10 and CIFAR-100).
Challenges: It does not help that an algorithm makes good predictions in 99% of cases if the 1% bad predictions are critical. To device efficient machine learning models that can say “I don’t know” is therefore both a business- and research challenge. Techniques such as ensembling and dropout can be used to quantify uncertainty in predictions, but can impose an increased computational burden at training and test time. To adopt AI in high-risk industries, uncertainty needs to be handled efficiently. We would like algorithms that are aware of their own weaknesses and can give us an indication when they are uncertain. They should say if the input looks very different from anything in the training data (e.g. detect distributional shifts). Ideally they should quantify evidence both for and against each class in the training set. Preferably they should also be able to distinguish epistemic uncertainty (uncertainty due to lack of knowledge, which can in principle be reduced by a better model) from aleatoric uncertainty (uncertainty because of actual variation in the world, which would remain even if the model was perfect). The student is encouraged to describe these challenges and desired features in more detail and to propose and explore possible solutions.
This project is performed in cooperation with DNV. DNV is a risk management and assurance company. This means that they help their customers across various industries to make sure that their products and processes are safe, sustainable and compliant with regulations. The Future of Digital Assurance program in DNV´s R&D department is tasked to explore new digital technologies that DNV can use to provide better assurance or may need to assure in the future. This includes the use and assurance of AI-based systems.
Supervision at IDI / NTNU: Prof. Rudolf Mester
Description: Deep learning has been reported to improve upon previous state of the art in many traditional machine learning tasks, like image classification, recommender systems, text-to-speech, and so on. Nevertheless, there are still fundamentally problematic issues with these systems, that invite theoretical work on extensions of deep learning towards (traditional) probabilistic reasoning. This is a topic of some interest, which has lead to nice tools that can be used for implementation/evaluation, like Tensorflow Probability and Pyro (built on top of PyTorch). For typical research trends, see Part III of "The Deep Learning Book" by Goodfellow et al. The selection of interesting research question(s) in this area will depend on the students' interest.
Level of sophistication: This is a difficult yet rewarding topic to work on. Students taking on this challenge should expect to do more demanding work than your run-of-the-mill project/master thesis. A good grasp of linear algebra, standard statistics, general machine learning, and deep learning is required for this to be fun. Furthermore, you will be expected to implement the ideas you develop in a deep learning framework, hence you will need to learn how to find your way around in Tensorflow, PyTorch, or similar, too.
The ongoing digital transformation of organisations and businesses along with the advancing convergence of information technology (IT) and operational technology (OT), provide significant operational challenges in terms of ensuring the stability, robustness, and security of these complex systems. As critical services become reliant on Internet-facing software systems it also makes them more vulnerable for cyberattacks. Recent studies also confirms that the number of targeted attacks is increasing, and that more sophisticated methods are taken into use.
At the same time, several voices are expressing their concern about whether current cybersecurity practices are sufficient to address these challenges [1]. A few scholars have recently started to look into how a resilience engineering perspective adopted from safety research could be used as a complementary perspective to understand the socio-technical nature of IT operations and incident response [2, 3]. This project will further study the current approaches to incident management of IT systems and aim to start untangling some of the collaborative practices that are central to sustaining software system performance.
[1] Bochman, A. The End of Cybersecurity. Harward Business Review. 2019. [2] Woods D.D. and Allspaw, J. 2020. Revealing the Critical Role of Human Performance in Software: It’s time to revise our appreciation of the human side of Internet-facing software systems. 2019. https://doi.org/10.1145/3380774.3380776 [3] Grøtan, T.O., Antonsen, S., Haavik, T.K. Cyber Resilience: A Pre-Understanding for an Abductive Research Agenda. 2022. https://doi.org/10.1007/978-3-030-85954-1_12
This thesis is assigned to Mia B.
Software runs the world and should be made for all people by all people. However, the softwareengineering task force is not gender balanced and software products exhibit gender biases. In 2023,merely 5.17% of the worldwide software developer population (27 millions) is comprised of women[1], [2]. Studies reveal a concerning trend: women in IT leave at a higher rate than men, with 50%resigning from tech roles before the age of 35 [3]. Many efforts have been devoted to understand why and how to foster diversity in software engineering [4] and how to mitigate the gender gap [5] [6] [7] [8] [9].The gender gap in science and technology has been attributed to several realities: Many individualsstill associate science and technology with masculine qualities, leading to stereotypes that candiscourage girls and women from pursuing education and careers in Science and Technology [10] [11].Unconscious biases in informing and tutoring young people and managing, promoting and grantingemployees can disadvantage women and lead to their low representation[12][13]. Women remain underrepresented in leadership positions, which makes it harder for girls and women to find role models and mentors in the field [14]. Careers can be demanding [15] [16], and some women may opt out or choose to work part-time to handle family responsibilities, which can impact career advancement.In 2023, global internet users reached 5.3 billion, constituting 65.7% of the world population [17].About 70% of males and 63% of females globally use the internet [18]. In Europe, only 20% ofICT specialists are female [19], and merely 5.17% of the worldwide software developer population (27 millions) is comprised of women [1], [2]. Studies reveal a concerning trend: women in IT leave at a higher rate than men, with 50% resigning from tech roles before the age of 35 [3].Gender-related studies within the open-source software community provide an overview of genderrepresentation, barriers to participation, and the experiences of women and non-binary individualsin open-source development [20]. OSS is also a source of data sets to investigate the relation between gender balance and quality of the software process [7] [21]. Researchers have investigated educational interventions to address gender disparities in computing classrooms [4] since late 90’s. Female participation in software engineering courses has historically been lower than desired, reflecting wider gender imbalances in the tech industry [22].The Research question to be addressed by this thesis will be: How do the career trajectories offemale software engineers evolve? The candidate will collect and analyze qualitative data by interviews and by social media (Linkedin) mining. The results will be a set of stories of career trajectories for the benefit of other researchers and of teachers and policy makers. Expected subjects for master theses 20 for PhD 50.References[1] Statista, Worldwide developer population, 2022. [Online]. Available: https://www.statista.com/statistics/627312/worldwide-developer-population/ (visited on 2023).[2] Statista, Worldwide developer gender, 20223. [Online]. Available: https://www.statista.com/statistics/1126823/worldwide-developer-gender/ (visited on 2023).[3] J. L. Glass, S. Sassler, Y. Levitte, and K. M. Michelmore, “What’s so special about STEM? Acomparison of women’s retention in STEM and professional occupations,” Social forces, vol. 92,no. 2, pp. 723–756, 2013.[4] A. Fisher, J. Margolis, and F. Miller, “Undergraduate women in computer science: Experience,motivation and culture,” ACM SIGCSE Bulletin, vol. 29, no. 1, pp. 106–110, 1997.1[5] D. G¨urer and T. Camp, “An ACM-W Literature Review on Women in Computing,” SIGCSEBull., vol. 34, no. 2, 121–127, 2002.[6] K. Albusays, P. Bjorn, L. Dabbish, D. Ford, E. Murphy-Hill, A. Serebrenik, and M.-A. Storey,“The diversity crisis in software development,” IEEE Software, vol. 38, no. 2, pp. 19–25, 2021.[7] G. Catolino, F. Palomba, D. A. Tamburri, A. Serebrenik, and F. Ferrucci, “Gender diversityand women in software teams: How do they affect community smells?” In 2019 IEEE/ACM41st International Conference on Software Engineering: Software Engineering in Society (ICSESEIS),IEEE, 2019, pp. 11–20.[8] K. Blincoe, O. Springer, and M. R. Wrobel, “Perceptions of gender diversity’s impact on moodin software development teams,” Ieee Software, vol. 36, no. 5, pp. 51–56, 2019.[9] L. Jaccheri, C. Pereira, and S. Fast, “Gender issues in computer science: Lessons learnt andreflections for the future,” in 2020 22nd International Symposium on Symbolic and NumericAlgorithms for Scientific Computing (SYNASC), IEEE, 2020, pp. 9–16.[10] S. A. Basow, Gender: Stereotypes and roles. Thomson Brooks/Cole Publishing Co, 1992.[11] D. W. Chambers, “Stereotypic images of the scientist: The draw-a-scientist test,” Science education,vol. 67, no. 2, pp. 255–265, 1983.[12] Y. Wang and D. Redmiles, “Implicit gender biases in professional software development: Anempirical study,” in 2019 IEEE/ACM 41st International Conference on Software Engineering:Software Engineering in Society (ICSE-SEIS), IEEE, 2019, pp. 1–10.[13] N. Imtiaz, J. Middleton, J. Chakraborty, N. Robson, G. Bai, and E. Murphy-Hill, “Investigatingthe effects of gender bias on github,” in 2019 IEEE/ACM 41st International Conference onSoftware Engineering (ICSE), IEEE, 2019, pp. 700–711.[14] S. Singh and D. Basu, “Impact on Women Undergraduate CS Students’ Experiences from aMentoring Program,” in Proceedings of the 52nd ACM Technical Symposium on Computer ScienceEducation, ser. SIGCSE ’21, New York, NY, USA: Association for Computing Machinery,2021, p. 1266.[15] E. Rubegni, B. Penzenstadler, M. Landoni, L. Jaccheri, and G. Dodig-Crnkovic, “Owning yourcareer paths: Storytelling to engage women in computer science,” in Gender in AI and Robotics:The Gender Challenges from an Interdisciplinary Perspective, Springer, 2023, pp. 1–25.[16] K. Sharma, J. C. Torrado, J. G´omez, and L. Jaccheri, “Improving girls’ perception of computerscience as a viable career option through game playing and design: Lessons from a systematicliterature review,” Entertainment Computing, vol. 36, 2021.[17] Statista, Digital population worldwide, 2023. [Online]. Available: https://www.statista.com/statistics/617136/digital-population-worldwide/ (visited on 2023).[18] Statista, Share of internet users worldwide by gender, 2022. [Online]. Available: https://www.statista.com/statistics/1362981/share-of-internet-users-worldwide-by-gender/(visited on 2023).[19] Eurostat, Worldwide developer gender, 2023. [Online]. Available: https://ec.europa.eu/eurostat/ (visited on 2023).[20] V. Singh and W. Brandon, “Open source software community inclusion initiatives to supportwomen participation,” in Open Source Systems: 15th IFIP WG 2.13 International Conference,OSS 2019, Montreal, QC, Canada, May 26–27, 2019, Proceedings 15, Springer, 2019, pp. 68–79.[21] M. Ortu, G. Destefanis, S. Counsell, S. Swift, R. Tonelli, and M. Marchesi, “How diverse is yourteam? investigating gender and nationality diversity in github teams,” PeerJ, Jul. 2016.[22] S. B. Berenson, K. M. Slaten, L. Williams, and C.-W. Ho, “Voices of women in a software engineeringcourse: Reflections on collaboration,” Journal on Educational Resources in Computing(JERIC), vol. 4, no. 1, 3–es, 2004.
The majority of modern applications are written in the so-called high-level productivity languages such as Python, NodeJS, Javascript, etc. In contrast, computer architecture and hardware research is mostly driven by software written in compiled languages such a C, C++ etc. The mismatch limits our understanding of how these applications are executed on the hardware/processors. For example, while the code written in C, C++ is handled by the “front-end” structures like instruction cache, branch predictors etc. of a processor, Python and NodeJS application code is handled by the “back-end” structures like data cache. This is because Python and NodeJS runtimes/interpreters are treated as code at hardware level, while both the application code and data are treated at data. Consequently our understanding of how to build efficient hardware/processors for the bulk of these applications is limited.
To achieve the level of understanding needed, we require better tools to measure the behaviour of such workloads throughout the computing pipeline. This project is concerned with designing and exploring the space for such a tool that can precisely pinpoint which cache lines contain applications data and which ones have the application code. While the tool helps us track the information in the front end, we must understand its effects on the “front-end” and “back-end” components. Understanding the impact of application behaviour on these components is of utmost importance to address their inefficiencies. Understanding their execution behaviour will allows us to propose new methods that not only ensure the efficient execution of such applications from performance perspective but also with regards to energy-efficiency.
The world beneath the waves holds a wealth of mysteries and potential discoveries. This project embarks on the journey of enhancing underwater exploration through object detection and classification techniques.
By harnessing sonar data, underwater imagery, and machine learning algorithms, the study aims to create an intelligent system capable of identifying and categorizing submerged objects with high accuracy. This research involves addressing the challenges posed by low visibility, varying environmental conditions, and diverse underwater ecosystems. The outcomes of this endeavor contribute to unlocking new insights into marine life, archaeological sites, and underwater infrastructure, ultimately aiding in the sustainable management and conservation of oceans.
Something we humans rarely notice, is how our eyes are in constant movement, and all of these constant movements can be put into four categories. Eye tracking is a technology used to capture an individual's eye movement, and is most commonly achieved by having a small infrared camera for each eye, and then use the center of the pupil as a starting point for further calculations to categorize the eye movements. Unfortunately, this categorization remains expensive, as it has to be conducted by specialists and is time consuming. An easily mistaken assumption in this regard, is that this is a task which would be trivial to automate by defining parameters for what makes a movement fall into a given category and simply use these algorithms to perform the classification. The reality is, however, different and considerably more complex. There are several reason for why this is more complicated that what one might initially assume, but inaccuracies in the captured data (e.g. due to hardware, pupil center algorithms, cornea reflections, etc.) is the most prominent one.
The thesis will be to map already existing, open, and annotated data sets, and to assess these from a quality perspective. Further, a look into how using different types of unsupervised learning, augmentation of the imagery, and other beneficial techniques/means, can benefit the accuracy of automatic annotation. A comparison of the annotations of the already existing and open datasets, to that of the automatic annotations performed, is then to be undertaken.
Additionally, it's desirable to explore the possibility of using the developed methods for eye tracking in VR headsets, as this introduces an extra dimension where the eye tracking cameras themselves will move slightly relative to the eye, due to the head movements which occur when moving about in VR.
This project aims to investigate the efficacy and practical considerations of the use of honeypots in cybersecurity defense strategies. Honeypots can give insights into attacker methodologies and in doing so can bolster overall network security. This project will explore the practical deployment of honeypots simulating n-day vulnerable software (e.g., vulnerable software with an outstanding CVE), focusing on their ability to detect and mitigate various types of cyber threats.
The preproject will involve conducting a systematic review following the PRISMA methodology on honeypots and their application to cybersecurity. The literature review will help the candidate(s) identify research gaps and areas of interest to focus on for the MSc thesis.
In this project we will consider how to use causal understanding to enhance a generative model, in order to obtain better generative processes. The initial idea, which of course is open for discussion, is something like the following:
Variational Auto-encoders (VAEs) are models that encode a high-dimensional datapoint (like an image) into a lower-dimensional code. Having access to the code, we can then decode this code again back to the original image. The power of this model comes from us being able to use it as a generator: Simply invent some code z ourselves, and decode that to generate a new data-point x. How then, should one make z to get an interesting object x? One well-explored solution is to ensure that the code-space is disentangled and hopefully also understandable. If one, for instance, work with a dataset of human faces, we would hope to see that one dimension of z represents the degree of which the person is smiling. We can now change the “smily-ness” of a person in a picture by encoding the image to get the code z, and simply change the code-dimension that controls smiling to suit our purpose (getting a new z') before decoding that manipulated code. This works fairly well, and has been around for some time (Example).
However, sometimes the different things that make up the code just are related. Say one dimension controls gender of the person depicted, another the length of the hair. If we were to intervene and change the gender of a depicted person (that is, change the gender-dimension), we may also accept or even want the system to adapt the length-of-hair accordingly. If we have such relationships stored in an extra knowledge representation, we may be able to make this happen. There are some attempts starting to explore this path (e.g., CausalVAE), and in this project we will explore these opportunities, and consider the quality of generation in light of the amount of side-information required to obtain the results.
Supervisors: Michail Giannakos (in collaboration with Ås Vitenparken)
Place: LCI Lab (https://lci.idi.ntnu.no/)
Suitable for: One or two students (2 recommended)
IntroductionScience-related subjects are some of the most influential subjects that children learn at an early age due to their ability to teach them to make observations, collect data, and come to conclusions logically. Skills like these are extremely valuable in everyday life. Additionally, research shows that children’s interest in science decreases or increases between the ages of 10 and 14 depending on their learning foundation and areas of interest. Some of the most frequent arenas for supporting children’s interest in science are science centers. Visits to science centers allow children to get to know different topics and scientific phenomena. Science centers are being benefited from mobile and interactive technologies, but it is unclear how different elements such as gamification and the use of analytics can support children’s interest. With this in mind, this thesis aims to investigate how interactive mobile technology (mobile application) enhanced with LLMs and conversational agents (e.g., open source examples CodeLLama, ParlAI, ChatterBot), as well as Multimodal LLMs can support children’s science learning.
Thesis DescriptionThe student(s) need to review the literature and familiarize themselves with relevant apps and AI technology. The focus is to integrate (M)LLM and conversational agent mechanics (e.g. allow children to ask questions during their visit) in an engaging way. Based on the best practices from the literature, the candidates will develop the app (e.g., following co-design or participatory design) and then do a user study either in school settings or in a science museum to empirically test the proposed system. Finally, the candidate will analyze the collected data and write up their thesis.
RequirementsThe ideal candidate will have a background in user experience, interface design, and use/integration of (M)LLMs. Solid front-end programming skills (JavaScript and CSS) and an interest in hands-on development and experimentation.
Programming skills: MySQL, JavaScript, CSS.
Examples of previous theses:https://ntnuopen.ntnu.no/ntnu-xmlui/handle/11250/3019915https://ntnuopen.ntnu.no/ntnu-xmlui/handle/11250/3019903
TL;DR: GitHub CodeQL is an analysis engine that can be used to perform queries on source code. We want to use CodeQL to write queries that verify coding rules for Java. Coding quality rules exist for different purposes, for example the SEI Cert Coding Standard focuses on security.
Some rules can be checked by static analysis tools, such as PMD [5] or SonarQube [6]. However, such tools only support a reduced number of rules. GitHub CodeQL [2] is an analysis engine that can be used to perform queries on source code. In CodeQL, code is treated like data. Security vulnerabilities, bugs, and other errors are modeled as queries that can be executed against databases extracted from code. The objective of this project is to use CodeQL to write queries that verify rules in the SEI Cert Oracle Coding Standard for Java [1], and then compare with existing static analysis tools. This work builds on a previous research work that is reported in [4].
The long-term research objective linked to this activity is to improve the process and techniques for the verification of code quality rules in source code.
Virtual devices (camera, microphone) are becoming increasingly interesting for producing visual and audio effects in digital meetings. This project will investigate the creation of a virtual camera and a virtual microphone in Windows (initially).
Following this, some cool applications will be created, such as voice change and changing facial characteristics.
For inspiration, some well-known packages (that will not be used) include OBS Virtual Camera and VB Cable:
https://obsproject.com/forum/resources/obs-virtualcam.539/
https://vb-audio.com/Cable/
Interested in Virtual, Augmented and Mixed Reality (i.e. XR) related to learning (e.g. AI/ML/DL) and training (e.g. medical cases)
PDF version of project proposal can be found here.
Mamba architectures, based on State-Space Models, have quickly generated a lot of interest due to their potential benefits over Convolutional Neural Networks (CNNs) and Transformers. Vision Mamba (ViM) offers the potential to model long-range dependencies effectively while maintaining computational advantages compared to Vision Transformer (ViT).
These potential benefits have catalyzed substantial research endeavors in 2024, investigating the feasibility of applying ViM to the medical imaging domain, as an alternative to Convolutional Neural Networks (CNNs) and ViT.
Project Goals
This thesis project investigates Vision Mamba's potential in medical image analysis. The goals are:
The balance between theory and practice will be tailored to the student's interests.
Relevant References
Note: The field of Vision Mamba in medicine is rapidly evolving. Incorporating the most recent publications and developments will be vital for a successful thesis.
Interested? Don't hesitate to contact either of:
Since its introduction in 2017, Transformers have revolutionised NLP and completely taken over most sequence processing.
From 2020 we have also seen examples of how transformers can be used for computer vision (often called Vision Transformers) and the number of publications related to this has almost exploded in early 2021, many of which outperforms current SotA in vaiours CV tasks.
In short this project will involve two main parts: 1) the theory (i.e. really understand how transformers / vision transformers work and investigate the latest works related to transformers for CV), and 2) the practical application of vision transformers in domains like Autonomous Vehicles / Robots and Medical Image Analysis (its also posible to suggest your own project related to vision transformers). The balance between theory and practice will be found in dialog with the interested student.
NLP: Attention Is All You Need (Transformers)
CV: An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (ViT)
You find an extensive list of Vision Transformer architectures here.
Understand and Explore the latest architectures for text to image / video / 3D for content generation (maybe in combination with AI gen. sound and music) for various applications.
Some examples:
A more concrete project, including potential application areas are decided together with students interested in the topic.
Task: Develop an app for mobile devices, that can be mounted in public transportation like busses, can access the camera of the device as well as relatively cheap and accurate positioning equipment (e.g. u-blox) with CPos corrections (cm accuracy) and have AI models for assessing and geo-referencing the condition of all road objects visible from the road (one application, other applications could be to create and update HD-maps, match real-time images to a reference for back-up localisation, collect data for neural rendering etc.).
Supervisors at NTNU: Frank Lindseth and Gabriel Kiss (drop us an email if you want to know more / discuss the project)
Visual SLAM is a term for a set of methods and algorithms that a) determine the motion of a camera (or a set of cameras) through an environment and b) determine the geometrical shape of that environment. vSLAM often builds on detecting “prominent points” in the images, and tracking them through the sequence. If a sufficient number of such points are tracked between two images, the relative pose (=translation and rotation) of the camera can be estimated. As any measurement in images is afflicted by errors, both these pose estimates as well as the estimated 3D positions of the observed image points are uncertain, and the estimation of the complete camera trajectory as well as the scene model “stitched together” from many views needs to be input data to a huge optimization problem.In AROS, we have access to both real video footage from underwater missions, as well as a realistic simulation environment which is able to generate video sequences where the motion and the 3D geometry are precisely known (‘ground truth’). The student project is integrated into our design and development process for a vSLAM system which is specifically tuned to be able with the substantial problems of underwater video material: limited visibility due to turbid water, bad illumination which is also moving with the robot vehicle, disturbances by plankton, dirt, and small fish, and many more. Which part of the vSLAM development is determined to be the focus area of the student project is subject to negotiation; the intention is to let the students experiment with novel approaches proposed in the recent literature, some of them focusing on geometric models and statistical estimation theory, others on machine learning. So we are able to adapt the topic largely to the background knowledge the student(s) already have, and their interest into different relevant research fields, such as e.g. state estimation, optimization, object detection and tracking, machine learning and deep learning.
Potential focus topics:* Robust keypoint tracking in the presence of underwater image degradation* Dynamic Model based prediction and correcting in keypoint and object tracking in underwater conditions* Pose graph and state sequence optimization for underwater visual SLAM* Integration of IMU measurements in underwater visual SLAM* Machine Learning for depth estimation, flow field estimation, and visual clutter detection
Literature:
D. Scaramuzza, F. Fraundorfer: Visual Odometry: Part I - The First 30 Years and Fundamentals. IEEE Robotics and Automation Magazine, 2011.F. Fraundorfer, D. Scaramuzza: Visual odometry: Part II - Matching, robustness, optimization, and applications. IEEE Robotics and Automation Magazine, 2012.
Cesar Cadena, Luca Carlone, et al.: Past, Present, and Future of Simultaneous Localization And Mapping: Towards the Robust-Perception Age. 2016
H. Zhan et al: DF-VO: What Should Be Learnt for Visual Odometry? 2021
Forskningssenter for nullutslippsområder i smarte byer (FME ZEN ) skal utvikle løsninger for framtidens bygninger og byområder, løsninger som bidrar til at nullutslippssamfunnet kan realiseres.Gjennom senteret vil kommuner, næringsliv, myndighetsorgan og forskere samarbeide tett for å planlegge, utvikle og drifte områder uten klimagassutslipp. Mer effektiv energibruk, produksjon og bruk av fornybar energi vil bidra til bedre miljøet lokalt og til å nå nasjonale klimamål.Oppgaven går ut på å utrede hvordan man skal følge opp kriterier for nullutslippsområder ved innsamling og visualisering av relevante data. Konkret vil det handle om utvikling av ulike dashboards knyttet til ulike områder, og den underliggende data-arkitekturen for å gi tilgang til nødvendige data.
Oppgaven vil følge en design science research metodikk, og besvarelsen forventes å skrives på engelsk. En god besvarelse bør kunne omgjøres til en vitenskaplig pulikasjon
Vi ansetter også forskningsassistenter/sommerjobber inn mot tematikken
In an ongoing collaboration with Klinikk for Fysikalsk Medisin og Rehabilitering Lian St. Olav, AIT/IDI is developing a novel VR environment for designing experiments to be used in research. Currently, the main focus is to allow neuropsychologists to design experiments with automatic data collection using multiple sensors. The masters students will initially look at integrating synchronized data from an EEG braincap and an eye tracker into an existing (fairly rudimentary) VR environment, and assess the feasibility of using this as a part of the overarching goal (with regards to accuracy, ease of use, visualization of the collected data, etc.)
Furthermore, the students will expand upon the existing environment by exploring which tools are necessary to increase the usefulness of the system. This can be achieved by focusing on a phenomenon called neglect, in which a patient following a stroke ignores part of their field of view. The traditional way to assess the severity of neglect is to use pen-and-paper tests, so implementing a corresponding suite of tests in VR is a possible avenue to explore the system's usefulness while simultaneously uncovering opportunities to collect data previously unavailable using the pen-and-paper tests (for example with the use of eye tracking and/or EEG).
The findings and results from this work is intended to be built upon by other students and researchers.
When analysts at ANEO A/S are planning wind-power production, they have to rely on the weather forecast. While in general of very high quality, the forecast of wind from the Norwegian Meteorological Institute also has some limitations:
In this project we want to explore machine learning for wind-production. The long-term goal is to have a highly localized weather forecast with at least 15min prediction frequency and explicit representation of uncertainty. To achieve this long-term goal we are now looking at several different directions, of which you can choose to pursue one:
This project will be co-supervised by Gleb Sizov and Odd Erik Gundersen (ANEO A/S) and Jørn Kristiansen (The Norwegian Meteorological Institute).
Over the last years the emergence of key technologies such as big data analytics and artificial intelligence have given rise to a completely new set of skills that are needed in private and public organizations. With IT gaining an increasingly central part in the shaping of business strategies, it is important that study curricula follow these requirements and provide graduates that fit the needs of organizations. This project will be run in collaboration with the Big Data Observatory (https://www.observatory.no) and involve collecting data through focus groups and surveys with key representatives. The output will involve a detailed look at what skills are necessary and how they can be addressed by educational institutions.
While there has been a lot of focus on the technical aspects related to artificial intelligence, recent years have seen a growing discussion about what the application of AI could be for private and public organizations. The objective of this master thesis project is to examine the readiness of private and public organizations to adopt AI, and the value they have derived from such investments. This project will involve collecting and analyzing data in collaboration with the researchers from the Big Data Observatory (https://www.observatory.no). It is an exciting opportunity to see how organizations are planning to use AI and what steps they need to take to adopt such technologies.
The job market, in Norway and internationally, has changed considerably over the past few years due to the COVID-19 pandemic and the emerging AI technologies, raising the need for developing innovative methods for workplace training and career guidance. In this project we will investigate how the use of Virtual Reality technologies and gaming elements can 1) motivate and inform young job seekers on their way to work and 2) contribute to faster skill acquisition for new employees. Through the simulation of a workplace or an industry (e.g. aquaculture or a shipyard), the job seekers can immerse into different workplaces and try out typical tasks, for example, salmon feeding or welding in a safe setting, thus mastering the corresponding real world situation.The master project will be performed in collaboration with Erasmus+ VR4VET project (Virtual Reality for Vocational Education and Training, https://vr4vet.eu/) involving several partners in Norway, Germany and Netherlands. The project proposes a new approach to vocational training and career guidance applying VR to allow active and engaging exploration of professions and introductory training, involving job seekers, career counsellors and industry stakeholders all over Europe. The student(s) will work in close collaboration with NAV, local industries (especially maritime) and our European partners (TU Delft and TH Köln). VR4VET is a continuation of Virtual Internship project that has so far resulted in several prototypes for workplace training and job interview training in VR and received international recognition (e.g. Best Demo Award at EuroVR 2018 and Breakthrough Auggie Award finalist) and broad media coverage https://memu.no/artikler/gir-ungdom-en-virtuell-jobbsmak/, https://www.ntnu.edu/imtel/virtual-internship.
The concept of e-governance proposes the use of ICT technologies towards operation and outreach. Extended-Reality technologies has been proposed as a tool to capture citizen perception towards strategies and policies. The question of “How” these technologies can be utilized for the most efficient transfer of knowledge, communication and capturing feedback remain unanswered to a high extent. Subjective constraints such as personal beliefs and experiences further complicate the problem of objectifying citizen output. Hence, the aim of the research is to build tools that can be used for efficient citizen engagement.
- XR application for citizen engagement
- Conceptualization of citizen engagement with XR
Augmented Reality (AR) can provide rich and interactive learning experiences and performance augmentation for remote distributed learners. The use of AR for collaboration and learning has significantly increased during the ongoing pandemic and has been widely adopted by several companies (e.g. Equinor), hospitals treating COVID-19 patients (https://www.businessinsider.com/london-doctors-microsoft-hololens-headsets-covid-19-patients-ppe-2020-5?r=US&IR=T) and educational institutions.
The goal of this master project is to perform research on the design principles and tools for AR-supported collaborative learning while working with several Hololens 2 units (https://www.microsoft.com/en-us/hololens/). Depending on the interests of the student(s), the project will be connected to a company (e.g. Equinor), St. Olavs hospital or a course taught at NTNU. The project is done in collaboration with IMTEL lab (https://www.ntnu.edu/ipl/imtel).
Supervisors: Gabriel Kiss (COMP/IDI), Ekaterina Prasolova-Førland (IMTEL/IPL)
XR in a teaching scenario can provide students an immersive, interactive experience that cannot be achieved in the real world. As such it can be provide rich and interactive learning experiences for local and distributed learners. Therefore, XR can be a useful tool for teaching computer graphics and computer science algorithms but also it has been widely used in the medical field. The expensive, bulky simulators can be replaced by VR/AR tools that provide a realistic experience. Furthermore, combining 3D printed artifacts or existing anatomic models and augmenting them with virtual elements has been proved a very effective teaching tools. Possible topics:- Virtual university, merging existing tools in a joint app, tools for teaching computer graphics / deep learning concepts in VR- Teaching echocardiography- Teaching fetal ultrasound acquisitions- Tools for understanding the relationship between heart function and cerebral blood flow in neonates (https://cimonmedical.com/neodoppler/)- Collaborative learning in AR/VR
The aim of this work is to extend a visualization framework using mixed reality, which may improve the way data is presented to an operator during a cardiac ultrasound exam or bronchoscopy procedure.One of the main challenges for an operator during the procedure is to process the data coming from various image sources displayed on several screens scattered around the room and combine this with apriori knowledge in the form of segmented anatomic models.By gathering information from important image sources using an existing open source research platform and presenting the data in an intelligent manner and then visualizing it in the operator's field of view, we hope to improve the ergonomic condition of the operator and increase the success rate of various procedure.Possible topics:- Bronchoscopy related- Echocardiography related
The projects I have made available are simply proposals for you to consider. However, the best projects come from motivated students, and if you have your own ideas regarding a project that is related to
please come and discuss your ideas with me.
Åpne data (sammen med dataplattformer) innebærer innsamling og deling av data på tvers av bransjer. Motivasjonen er at ved å gjøre data åpent tilgjengelig, økes produktiviteten gjennom økt samarbeid eller skaper mer velfungerende markeder. Et vellykket eksempel er hvordan legemiddelindustrien samarbeidet om å utvikle og distribuere Covid-vaksiner i rekordfart. Andre eksempler er: EUs PSD2-regulering mot open banking i finans og BarentsWatch for overvåking av havene.
Åpne data er nøkkelen til regjeringens digitaliseringsstrategi, illustrert i denne ferske stortingsmeldingen, https://www.regjeringen.no/no/dokumenter/meld.-st.-22-20202021/id2841118/?ch=5
Utfordringen er at det i dag er sosiale, tekniske, praktiske og organisatoriske forhold som må være på plass for at visjoner om åpne data skal vokse frem i praksis. Eksempler på utfordringer er motivasjon, gjennomførbarhet av teknisk infrastruktur, belønninger og betalingsmodeller.
Caset i denne oppgaven er fiskeri. Fiskeri er en av norges viktigste næringer, sentral i fornybarsatsing og noe Norge skal kunne leve av lenge etter vi er ferdige med fossile næringer. Fiskerinæringen er høyteknologisk og tar stadig i bruk nye teknologier, som sensorer, dataplattformer, åpne data og kunstig intelligens (f.eks. bildegjenkjenning).
Fiskeridirektoratet i Norge søker å ta i bruk nye teknologiske løsninger som kunstig intelligens og bildegjenkjenning for å sikre korrekt registrering av fangst (https://www.fiskeridir.no/Yrkesfiske/fangstid). Noen av fordelene med registrering kan være at det bidrar til å forhindre fiskerikriminalitet og konkurransevridning. I tillegg støtter det et bærekraftig fiske. Imidlertid er næringen, slik som fiskere og rederier, skeptiske da det oppleves som (nok) et ledd i økt myndighetskontroll.
Oppgaven vil gjøres i samarbeid med SINTEF Nord, Norges Fiskerihøgskole (UiT Norges arktiske Universitet), fiskeridirektoratet og fiskerinæringen.
