Cooperative Multi-Agent Reinforcement Learning with Hypergraph Convolution

TL;DR

This paper introduces HGCN-MIX, a hypergraph convolution-based method for multi-agent reinforcement learning that improves coordination among agents, especially in large-scale scenarios, by modeling their relationships explicitly.

Contribution

It proposes a novel hypergraph convolution approach integrated with value decomposition to enhance agent coordination in multi-agent systems.

Findings

HGCN-MIX outperforms current state-of-the-art methods in StarCraft II benchmarks.

The method shows greater improvements with more agents involved.

Learning more relationships in the hypergraph leads to stronger joint policies.

Abstract

Recent years have witnessed the great success of multi-agent systems (MAS). Value decomposition, which decomposes joint action values into individual action values, has been an important work in MAS. However, many value decomposition methods ignore the coordination among different agents, leading to the notorious "lazy agents" problem. To enhance the coordination in MAS, this paper proposes HyperGraph CoNvolution MIX (HGCN-MIX), a method that incorporates hypergraph convolution with value decomposition. HGCN-MIX models agents as well as their relationships as a hypergraph, where agents are nodes and hyperedges among nodes indicate that the corresponding agents can coordinate to achieve larger rewards. Then, it trains a hypergraph that can capture the collaborative relationships among agents. Leveraging the learned hypergraph to consider how other agents' observations and actions affect their decisions, the agents in a MAS can better coordinate. We evaluate HGCN-MIX in the StarCraft II multi-agent challenge benchmark. The experimental results demonstrate that HGCN-MIX can train joint policies that outperform or achieve a similar level of performance as the current state-of-the-art techniques. We also observe that HGCN-MIX has an even more significant improvement of performance in the scenarios with a large amount of agents. Besides, we conduct additional analysis to emphasize that when the hypergraph learns more relationships, HGCN-MIX can train stronger joint policies.