Multi-Agent Continuous Control with Generative Flow Networks

TL;DR

This paper introduces MACFN, a multi-agent generative flow network approach that enables cooperative exploration in continuous control tasks by decomposing global flows into local agent contributions, outperforming existing methods.

Contribution

The paper proposes a novel decentralized multi-agent flow decomposition method with theoretical guarantees, enhancing exploration in continuous control environments.

Findings

MACFN outperforms state-of-the-art methods in continuous control tasks.

The flow decomposition network effectively assigns local flows to agents.

Experimental results show improved exploration capabilities.

Abstract

Generative Flow Networks (GFlowNets) aim to generate diverse trajectories from a distribution in which the final states of the trajectories are proportional to the reward, serving as a powerful alternative to reinforcement learning for exploratory control tasks. However, the individual-flow matching constraint in GFlowNets limits their applications for multi-agent systems, especially continuous joint-control problems. In this paper, we propose a novel Multi-Agent generative Continuous Flow Networks (MACFN) method to enable multiple agents to perform cooperative exploration for various compositional continuous objects. Technically, MACFN trains decentralized individual-flow-based policies in a centralized global-flow-based matching fashion. During centralized training, MACFN introduces a continuous flow decomposition network to deduce the flow contributions of each agent in the presence of only global rewards. Then agents can deliver actions solely based on their assigned local flow in a decentralized way, forming a joint policy distribution proportional to the rewards. To guarantee the expressiveness of continuous flow decomposition, we theoretically derive a consistency condition on the decomposition network. Experimental results demonstrate that the proposed method yields results superior to the state-of-the-art counterparts and better exploration capability. Our code is available at https://github.com/isluoshuang/MACFN.