Learning and Computational Game Theory:
Special Issue of the Machine Learning Journal

Amy Greenwald and Michael L. Littman, guest editors

Game theory is concerned with the decisions made by utility-maximizing individuals in their interactions with other decision makers and their environment. From its earliest days of study, researchers have recognized the important relationship between game theory and learning---using experience from past play to guide future decisions. Recently, there has been a tremendous increase in research that applies a computational perspective to learning and games. Exciting results showing, for example, convergence of learning algorithms to game-theoretic equilibria and computationally efficient representations for strategic interactions, are helping to shape our understanding of learning in the multiagent setting.

The purpose of this special issue on learning in games is to give active researchers an opportunity to share significant contributions in this rapidly growing area at the intersection of machine learning and economics.

The emphasis of the special issue will be on how decision makers with individual utility functions learn to behave in interactive situations. It is critical that all submissions clearly state the problem setting in which the results apply---what information is available to the individual players (their own actions, the actions of others, utilities of all players, identity of the other players, etc.), how their performance is to be judged (utility of the player, social utility, convergence to equilibrium, stability of learned behavior, etc.), the model of uncertainty (randomized payoffs, noise in action perception, stochastic action effects, etc.), the information structure of the game and the permissible equilibrium concepts, and any computational complexity concerns.

Submissions are expected to represent high-quality, significant contributions in the area of machine learning and computational game theory. Authors are encouraged to follow formatting guidelines for Machine Learning manuscripts.

Administrative notes

Schedule

Submission Deadline: September 1, 2005
Send Papers to Reviewers:September 15, 2005
Reviews Due Back to Editors:November 1, 2005
Decisions Announced: November 15, 2005
Camera-Ready Due: January 15, 2006
Print Publication: early 2006

This page maintained by Michael Littman mlittman@cs.rutgers.edu