Model-based Multi-agent Policy Optimization with Adaptive Opponent-wise Rollouts

Abstract

This paper investigates the model-based methods in multi-agent reinforcement learning (MARL). We specify the dynamics sample complexity and the opponent sample complexity in MARL, and conduct a theoretic analysis of return discrepancy upper bound. To reduce the upper bound with the intention of low sample complexity during the whole learning process, we propose a novel decentralized model-based MARL method, named Adaptive Opponent-wise Rollout Policy Optimization (AORPO). In AORPO, each agent builds its multi-agent environment model, consisting of a dynamics model and multiple opponent models, and trains its policy with the adaptive opponent-wise rollout. We further prove the theoretic convergence of AORPO under reasonable assumptions. Empirical experiments on competitive and cooperative tasks demonstrate that AORPO can achieve improved sample efficiency with comparable asymptotic performance over the compared MARL methods.

Paper accepted at IJCAI 2021

References in corpus (13)

• Soft Actor-Critic Algorithms and Applications
• Searching for Activation Functions
• Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents
• Mean Field Multi-Agent Reinforcement Learning
• Benchmarking Model-Based Reinforcement Learning
• Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Control
• Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees
• Multiagent Soft Q-Learning
• Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning
• Model-Augmented Actor-Critic: Backpropagating through Paths
• Value Propagation for Decentralized Networked Deep Multi-agent Reinforcement Learning
• Multi-agent Trajectory Prediction with Fuzzy Query Attention
• Multi-Agent Reinforcement Learning with Multi-Step Generative Models