Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments

TLDR

Traditional multi‑agent RL struggles because Q‑learning faces non‑stationarity and policy gradients suffer increasing variance with more agents. The study investigates deep reinforcement learning techniques for multi‑agent environments. The authors adapt actor‑critic algorithms to account for other agents’ policies and employ an ensemble training regimen to produce robust multi‑agent policies. The proposed method outperforms existing approaches in both cooperative and competitive settings, enabling agents to discover diverse physical and informational coordination strategies.

Abstract

We explore deep reinforcement learning methods for multi-agent domains. We begin by analyzing the difficulty of traditional algorithms in the multi-agent case: Q-learning is challenged by an inherent non-stationarity of the environment, while policy gradient suffers from a variance that increases as the number of agents grows. We then present an adaptation of actor-critic methods that considers action policies of other agents and is able to successfully learn policies that require complex multi-agent coordination. Additionally, we introduce a training regimen utilizing an ensemble of policies for each agent that leads to more robust multi-agent policies. We show the strength of our approach compared to existing methods in cooperative as well as competitive scenarios, where agent populations are able to discover various physical and informational coordination strategies.