Game Theory in Computer Science

Module code: CO7212

Module co-ordinator: Dr Rob Van Stee

Modern computer science has to deal with large, heterogeneous networks in which a large number of autonomous agents interact. Often, such systems are not centrally planned but evolve in a distributed fashion as a result of the interaction of agents. These can be modelled using concepts from game theory.

In this module you will study the basic concepts of game theory and discuss their use in the solution and modelling of problems faced by computer scientists. We will consider the prediction of the equilibrium state reached via the interaction of selfish users in a communication network, and the comparison of that state with a globally optimised state. We will also look at how mechanisms are designed to ensure that individual players will behave in a way that achieves a desirable global state of the system.

Learning

  • 24 one-hour lectures
  • 8 one-hour seminars
  • 8 one-hour tutorials

Assessment

  • Exam, 2 hours (60%)
  • Coursework (40%)