Learning Policy Representations in Multiagent Systems

TL;DR

This paper introduces a general unsupervised learning framework for modeling agent behavior in multiagent systems, using minimal interaction data to learn policy representations applicable across diverse environments.

Contribution

It presents a novel representation learning approach for agent modeling that is task-agnostic and does not rely on domain-specific prior knowledge.

Findings

Effective in high-dimensional competitive environments

Successful in cooperative communication tasks

Improves policy optimization in reinforcement learning

Abstract

Modeling agent behavior is central to understanding the emergence of complex phenomena in multiagent systems. Prior work in agent modeling has largely been task-specific and driven by hand-engineering domain-specific prior knowledge. We propose a general learning framework for modeling agent behavior in any multiagent system using only a handful of interaction data. Our framework casts agent modeling as a representation learning problem. Consequently, we construct a novel objective inspired by imitation learning and agent identification and design an algorithm for unsupervised learning of representations of agent policies. We demonstrate empirically the utility of the proposed framework in (i) a challenging high-dimensional competitive environment for continuous control and (ii) a cooperative environment for communication, on supervised predictive tasks, unsupervised clustering, and policy optimization using deep reinforcement learning.