Variational Autoencoders for Opponent Modeling in Multi-Agent Systems

TL;DR

This paper introduces a variational autoencoder-based approach for opponent modeling in multi-agent systems, enabling an agent to learn and adapt to fixed-policy opponents using only local information, without access to opponents' actions or observations.

Contribution

The paper proposes a novel opponent modeling method using variational autoencoders that does not require access to opponents' actions or observations during training or execution.

Findings

Achieves equal or higher episodic returns compared to existing methods.

Effective opponent modeling with only local information.

Improves agent performance in multi-agent reinforcement learning tasks.

Abstract

Multi-agent systems exhibit complex behaviors that emanate from the interactions of multiple agents in a shared environment. In this work, we are interested in controlling one agent in a multi-agent system and successfully learn to interact with the other agents that have fixed policies. Modeling the behavior of other agents (opponents) is essential in understanding the interactions of the agents in the system. By taking advantage of recent advances in unsupervised learning, we propose modeling opponents using variational autoencoders. Additionally, many existing methods in the literature assume that the opponent models have access to opponent's observations and actions during both training and execution. To eliminate this assumption, we propose a modification that attempts to identify the underlying opponent model using only local information of our agent, such as its observations, actions, and rewards. The experiments indicate that our opponent modeling methods achieve equal or greater episodic returns in reinforcement learning tasks against another modeling method.