TUTORIAL DESCRIPTION

TA1: Knowledge representation for game theory

**Tutorialists: Michael Wooldridge**

Saturday, August 3rd, morning

If we aim to build artificial agents that must act in environments populated by other artificial agents, then it is very natural to ask what mathematical models and concepts we can use to understand the interactions that might take place between such agents. Game theory is one answer to this question. Game theory is a branch of micro-economics which attempts to model and understand rational interaction between self-interested agents. This tutorial is aimed at students who want to learn what game theory is about, why it is relevant for AI researchers in general and multi-agent systems researchers in particular. Assuming no prior knowledge of game theory, the tutorial starts by motivating the role of game theory in AI, introduces the student to the key concepts of game theory (both non-cooperative and cooperative), the key solution concepts for non-cooperative games, and the various game theoretic models used by AI researchers.