A near-optimal polynomial time algorithm for learning in certain classes of stochastic games

We present a new algorithm for polynomial time learning of optimal behavior in single-controller stochastic games. This algorithm incorporates and integrates important recent results of Kearns and Singh (Proc. ICML-98, 1998) in reinforcement learning and of Monderer and Tennenholtz (J. Artif. Intell...

Full description

Saved in:
Bibliographic Details
Published in:Artificial intelligence 2000-08, Vol.121 (1), p.31-47
Main Authors: Brafman, Ronen I., Tennenholtz, Moshe
Format: Article
Language:English
Subjects:
Algorithms
Artificial intelligence
Exploration versus exploitation in multi-agent systems
Polynomial time learning in hostile environments
Stochastic games
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a new algorithm for polynomial time learning of optimal behavior in single-controller stochastic games. This algorithm incorporates and integrates important recent results of Kearns and Singh (Proc. ICML-98, 1998) in reinforcement learning and of Monderer and Tennenholtz (J. Artif. Intell. Res. 7, 1997, p. 231) in repeated games. In stochastic games, the agent must cope with the existence of an adversary whose actions can be arbitrary. In particular, this adversary can withhold information about the game matrix by refraining from (or rarely) performing certain actions. This forces upon us an exploration versus exploitation dilemma more complex than in Markov decision processes in which, given information about particular parts of a game matrix, the agent must decide how much effort to invest in learning the unknown parts of the matrix. We present a polynomial time algorithm that addresses these issues in the context of the class of single controller stochastic games, providing the agent with near-optimal return.
ISSN:0004-3702
1872-7921
DOI:10.1016/S0004-3702(00)00039-4