All learning is Local: Multi-agent Learning in Global Reward Games

Abstract

In large multiagent games, partial observability, coordination, and credit assignment persistently plague attempts to design good learning algo-rithms. We provide a simple and efficient algorithm that in part uses a linear system to model the world from a single agent’s limited perspec-tive, and takes advantage of Kalman filtering to allow an agent to con-struct a good training signal and effectively learn a near-optimal policy in a wide variety of settings. 1