Representation and Learning in Computational Game Theory

This NSF-sponsored project seeks to expand our understanding of how computational learning theory and rich representation can be used in the context of strategic interaction of self-interested agents.

People: co-PIs

Michael Kearns, University of Pennsylvania: Institute for Research in Cognitive Science.
Michael L. Littman, Rutgers University: Rutgers Laboratory for Real-Life Reinforcement learning.
Robert E. Schapire, Princeton University.
Manfred K. Warmuth, University of California, Santa Cruz.

People: Students

David LeRoux, Rutgers
Alexander L. Strehl, Rutgers
Cynthia Rudin, Princeton

Project Publications

A polynomial-time Nash equilibrium algorithm for repeated games. Michael L. Littman and Peter Stone. Special Issue of Decision Support Systems on the Fourth ACM Conference on Electronic Commerce. In press, 2004.

Related Work from Other Labs

How to combine expert (or novice) advice when actions impact the environment. Daniela Pucci de Farias and Nimrod Megiddo. NIPS 2004.