The players in a prisoners dilemma “defect” and receive less benefit compared to what they would get if they had cooperated. However, if they play the game many times (and particularly if they didn’t know how many times) then cooperation may emerge. So, first iterated games have other dynamics than the one-shot games. Second, N-player games are different than 2-player games. Finally, the games are radically different in nature where mechanisms that may emerge cooperation in one play don’t work in another game. These said, this project focuses on “public good games” (e.g., snowdrift game) which involves a contradiction between the rational behavior of individuals and the optimum outcome for the society. Despite this contrast and expectations of non-cooperative behavior according to traditional game theory, in real life cooperation has been observed to emerge. This has been attained to mechanisms such as kinship or close network of people which involve direct or indirect reciprocity. This project will have a focus on the role of reputation in promoting cooperation. Evolutionary game theory methods will be used in simulation and evaluation of designed reputation-based incentive mechanisms. 1-2 students
Det er en fordel å ha tatt Bio-inspirerte metoder faget (IT 3708).
Co-supervisor: Associate prof. Hai Nguyen www.ntnu.edu/employees/hai.nguyen