NTNU Norges teknisk-naturvitenskapelige universitet
  Fakultet for informasjonsteknologi, matematikk og elektroteknikk > Institutt for datateknikk og informasjonsvitenskap

IT3105 - Kunstig intelligens programmering, høst 2017

Contents
  • Home

  • Schedules

  • Projects

  • Grading

  • Lectures

  • Materials

  • News

  • Results


  • Grading, Delivery and Attendance Policies

    There is no final exam in this year's (autumn 2017) version of the course. Rather, a student's final grade will be based exclusively on the grades awarded to each of the projects. The exact point breakdowns for each project will be determined during the course of the semester, though it is normally pretty evenly balanced between the 2-4 projects.

    This year (2017), a single instructor will run the entire course and grade or oversee the grading of all projects. Some years, however, each project may be graded by a different teacher. Although these instructors work together to provide a quality course, they do NOT grade in exactly the same way. Do not assume that one teacher's grading policy is identical to another's. When in doubt, ask each instructor about their policy.

    Individual and Group Work

    You can work on the projects alone or in groups of TWO, but no larger. That is, the total group size cannot be larger than 2.

    For group deliveries, the following protocol must be followed:

    • There is one demo session for the group, and each group produces a single (shared) report, but
    • when uploading to BLACKBOARD, each group member must upload the same zip file (typically containing code plus a report).
    • Uploads to BLACKBOARD must be completed prior to the beginning of a demo session. If the day's first demos are at 9 am, then ALL students in ALL groups will have to do their uploads before 9 am.

    Many years of experience with this and related courses indicate that there is no obvious advantage nor disadvantage to working alone versus in a pair. Good programmers often like to work alone, and, in the past, the results have been just as good as those produced by groups. The instructors do not base project size on any assumptions that people will or will not work together.

    You are free to discuss problems with your classmates, but keep the discussions at a high level. All students/groups must produce the working code and the report for the projects by themselves and should be able to answer basic questions about any portion of their own (or the group's) code. Any obvious inabilities of group members to explain their code can result in significant point loss on a project.

    Downloading and Sharing Code

    Direct copying of any part of a project can result in an immediate failing mark (F) for the entire course. This pertains to the copier as well as the source!

    In some cases, it is acceptable to download working code for PERIPHERAL aspects of a project. For the AI projects of this course, the graphical user interface (GUI) is often considered peripheral. Such code can also be shared among students in the class. HOWEVER, it is strongly recommended that you consult the course instructor in any case where the acceptability of copying and sharing could be in doubt. Typically, all AI aspects of a project are NOT considered peripheral and should NOT be copied nor shared. In addition, many non-AI aspects of these projects are complex and meant to be challenging programming tasks; they too must be coded by each student or 2-person group.

    Late Deliveries

    In most cases, late deliveries are NOT accepted in this course. Although partial credit is often given on assignments, it is NOT given for late deliveries. There are exceptions to this (very strict) rule, and these typically involve serious or long-lasting personal illness (with a doctor's note) or family tragedy (e.g. a funeral). Happy events such as birthdays, weddings and vacations are normally known well in advance, so you will need to plan your work (and delivery) schedule around them. Having a lot of work in other classes is not a legitimate excuse for a late delivery.

    Early deliveries can occasionally be negotiated with the instructor or the course assistants, but these are also granted sparingly. Demo sessions are a lot of work for everyone involved in this course; adding extra sessions for one or a few students is generally very inconvenient.

    Course Attendance

    Students are not required to attend lectures, although their content is primarily designed to assist students with their projects. Most, but not necessarily all, lecture materials will be available online, but reading lecture slides on your own is normally less instructive than listening to their presentation.

    However, Demo session attendance is mandatory .

    RedaktÝr: Kontorsjef: Eivind Voldhagen  Kontaktadresse: Audun Liberg  Sist oppdatert: 01.01.1970