Context-Aware Sparse Deep Coordination Graphs

TL;DR

This paper introduces a novel method for constructing context-aware sparse coordination graphs in multi-agent systems, leveraging payoff function variance and action representations to improve coordination efficiency and robustness.

Contribution

It proposes a new variance-based approach for adaptive sparse graph construction and incorporates action representation learning to enhance multi-agent coordination.

Findings

The method effectively reduces the likelihood of action changes after edge removal.

Empirical results demonstrate improved coordination performance on MACO and StarCraft II benchmarks.

The approach provides insights into the dynamics of sparse graph learning in multi-agent systems.

Abstract

Learning sparse coordination graphs adaptive to the coordination dynamics among agents is a long-standing problem in cooperative multi-agent learning. This paper studies this problem and proposes a novel method using the variance of payoff functions to construct context-aware sparse coordination topologies. We theoretically consolidate our method by proving that the smaller the variance of payoff functions is, the less likely action selection will change after removing the corresponding edge. Moreover, we propose to learn action representations to effectively reduce the influence of payoff functions' estimation errors on graph construction. To empirically evaluate our method, we present the Multi-Agent COordination (MACO) benchmark by collecting classic coordination problems in the literature, increasing their difficulty, and classifying them into different types. We carry out a case study and experiments on the MACO and StarCraft II micromanagement benchmark to demonstrate the dynamics of sparse graph learning, the influence of graph sparseness, and the learning performance of our method. (The MACO benchmark and codes are publicly available at https://github.com/TonghanWang/CASEC-MACO-benchmark.)