POWQMIX: Weighted Value Factorization with Potentially Optimal Joint Actions Recognition for Cooperative Multi-Agent Reinforcement Learning

TL;DR

POWQMIX introduces a weighted value factorization method that recognizes potentially optimal joint actions, improving policy learning in cooperative multi-agent reinforcement learning environments.

Contribution

It proposes a novel weighted QMIX algorithm that overcomes monotonicity constraints, with theoretical guarantees for recovering the optimal policy.

Findings

Outperforms state-of-the-art methods in various environments

Proves theoretical guarantee of optimal policy recovery

Demonstrates effectiveness in complex multi-agent tasks

Abstract

Value function factorization methods are commonly used in cooperative multi-agent reinforcement learning, with QMIX receiving significant attention. Many QMIX-based methods introduce monotonicity constraints between the joint action value and individual action values to achieve decentralized execution. However, such constraints limit the representation capacity of value factorization, restricting the joint action values it can represent and hindering the learning of the optimal policy. To address this challenge, we propose the Potentially Optimal Joint Actions Weighted QMIX (POWQMIX) algorithm, which recognizes the potentially optimal joint actions and assigns higher weights to the corresponding losses of these joint actions during training. We theoretically prove that with such a weighted training approach the optimal policy is guaranteed to be recovered. Experiments in matrix games, difficulty-enhanced predator-prey, and StarCraft II Multi-Agent Challenge environments demonstrate that our algorithm outperforms the state-of-the-art value-based multi-agent reinforcement learning methods.