Playing is believing: The role of beliefs in multi-agent learning

We propose a new classification for multi-agent learning algorithms, with each league of players characterized by both their possible strategies and possible beliefs. Using this classification, we review the optimality of existing algorithms, including the case of interleague play. We propose an incremental improvementto the existing algorithms that seems to achieve average payoffs that are at least the Nash equilibrium payoffs in the longrun against fair opponents.

• Type: Conference
• Year: 2001
• Where: NIPS
• Authors: Yu-Han Chang, Leslie Pack Kaelbling
• Keywords: Average Payoffs, Multi-agent Learning Algorithms, Nash Equilibrium Payoffs, NIPS 2001, NIPS 2007