Locality Matters: A Scalable Value Decomposition Approach for Cooperative Multi-Agent Reinforcement Learning

TL;DR

This paper introduces LOMAQ, a scalable value decomposition algorithm for cooperative multi-agent reinforcement learning that leverages locality structures to improve scalability, performance, and convergence speed in large environments.

Contribution

The paper presents LOMAQ, a novel value-based multi-agent algorithm that incorporates local rewards and a reward decomposition method, enhancing scalability and efficiency in cooperative MARL.

Findings

LOMAQ scales better than existing methods in large environments.

LOMAQ significantly improves performance and convergence speed.

The reward decomposition method effectively finds local rewards from global signals.

Abstract

Cooperative multi-agent reinforcement learning (MARL) faces significant scalability issues due to state and action spaces that are exponentially large in the number of agents. As environments grow in size, effective credit assignment becomes increasingly harder and often results in infeasible learning times. Still, in many real-world settings, there exist simplified underlying dynamics that can be leveraged for more scalable solutions. In this work, we exploit such locality structures effectively whilst maintaining global cooperation. We propose a novel, value-based multi-agent algorithm called LOMAQ, which incorporates local rewards in the Centralized Training Decentralized Execution paradigm. Additionally, we provide a direct reward decomposition method for finding these local rewards when only a global signal is provided. We test our method empirically, showing it scales well compared to other methods, significantly improving performance and convergence speed.