Learning in Multiagent Systems: An Introduction from a Game-Theoretic Perspective

TL;DR

This paper provides an overview of learning in multiagent systems from a game-theoretic perspective, discussing key concepts, theories, and challenges in the field.

Contribution

It introduces the application of game theory to multiagent learning, including theories like fictitious play and replicator dynamics, and discusses engineering approaches like CLRI and n-level learning agents.

Findings

Theorems on fictitious play and replicator dynamics

Introduction of CLRI theory and n-level learning agents

Summary of remaining challenges in the field

Abstract

We introduce the topic of learning in multiagent systems. We first provide a quick introduction to the field of game theory, focusing on the equilibrium concepts of iterated dominance, and Nash equilibrium. We show some of the most relevant findings in the theory of learning in games, including theorems on fictitious play, replicator dynamics, and evolutionary stable strategies. The CLRI theory and n-level learning agents are introduced as attempts to apply some of these findings to the problem of engineering multiagent systems with learning agents. Finally, we summarize some of the remaining challenges in the field of learning in multiagent systems.