Coordinated Proximal Policy Optimization

Abstract

We present Coordinated Proximal Policy Optimization (CoPPO), an algorithm that extends the original Proximal Policy Optimization (PPO) to the multi-agent setting. The key idea lies in the coordinated adaptation of step size during the policy update process among multiple agents. We prove the monotonicity of policy improvement when optimizing a theoretically-grounded joint objective, and derive a simplified optimization objective based on a set of approximations. We then interpret that such an objective in CoPPO can achieve dynamic credit assignment among agents, thereby alleviating the high variance issue during the concurrent update of agent policies. Finally, we demonstrate that CoPPO outperforms several strong baselines and is competitive with the latest multi-agent PPO method (i.e. MAPPO) under typical multi-agent settings, including cooperative matrix games and the StarCraft II micromanagement tasks.

References in corpus (10)

• Continuous control with deep reinforcement learning
• Trust Region Policy Optimization
• High-Dimensional Continuous Control Using Generalized Advantage Estimation
• Counterfactual Multi-Agent Policy Gradients
• QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement Learning
• A Survey of Learning in Multiagent Environments: Dealing with Non-Stationarity
• Dealing with Non-Stationarity in Multi-Agent Deep Reinforcement Learning
• MAVEN: Multi-Agent Variational Exploration
• Towards Playing Full MOBA Games with Deep Reinforcement Learning
• Multi-Agent Trust Region Policy Optimization