OREGON STATE UNIVERSITY

You are here

Bayesian role discovery for multi-agent reinforcement learning

TitleBayesian role discovery for multi-agent reinforcement learning
Publication TypeConference Paper
Year of Publication2010
AuthorsWilson, A., A. Fern, and P. Tadepalli
Conference NameInternational Conference on Autonomous Agents and Multiagent Systems (AAMAS-10)
Pagination1587-1588
Date Published05/2010
Conference LocationToronto, Canada
Abstract

In this paper we develop a Bayesian policy search approach for Multi-Agent RL (MARL), which is model-free and allows for priors on policy parameters. We present a novel optimization algorithm based on hybrid MCMC, which leverages both the prior and gradient information estimated from trajectories. Our experiments demonstrate the automatic discovery of roles through reinforcement learning in a real-time strategy game.