CS825 - Game Theory and Multiagent Systems

Aims and Objectives

Students will be familiar with the mathematical theory of games and the ways in which it is applied to the study of multi-agent systems and in machine learning.

Learning Outcomes

After completing this module participants will be able to:

• understand the core mathematical concepts of game theory.
• know key examples of games and their properties.
• be able to use software tools for computing with games.
• understand how concepts of game theory are applied to the field of A.I.
• be able to build simple A.I. systems that play games.

Syllabus

• Utility functions, decisions under uncertainty.
• Normal form games, pure and mixed strategies and Nash equilibria, minmax, Pareto optimality, correlated equilibria.
• Extensive form games, subgame perfect equilibria, backward induction, information sets.
• Infinitely repeated games, the Folk theorem. Iterated prisoner’s dilemma.
• Bayesian games, Bayesian Nash equilibria.
• Markov games and reinforcement learning.
• Real world game play. Alpha-beta pruning. Tic-tac-toe, Chess, Go, StarCraft.
• Economic games and pricing algorithms.

Recommended Reading

This list is indicative only – the class lecturer may recommend alternative reading material. Please do not purchase any of the reading material listed below until you have confirmed with the class lecturer that it will be used for this class.

Essentials of game theory: A concise multidisciplinary introduction, Leyton-Brown, K. and Shoham, Y. Morgan & Claypool, 2008 | Andersonian Library (Other retailers are available)

Gambit documentation at https://gambitproject.readthedocs.io/en/latest/

Last updated: 2022-09-09 13:43:27