Interference-Aware K-Step Reachable Communication in Multi-Agent Reinforcement Learning

arXiv cs.AI / 3/17/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

IA-KRC introduces a new framework for interference-aware communication in multi-agent reinforcement learning to address limited bandwidth and dynamic topologies.
It combines a K-Step reachability protocol that confines message passing to physically accessible neighbors with an interference-prediction module that selects partners by minimizing interference and maximizing utility.
Compared with existing methods, IA-KRC achieves more persistent and efficient cooperation under environmental interference.
Comprehensive evaluations demonstrate superior performance, robustness, and scalability in complex, highly dynamic multi-agent scenarios.

Abstract

Effective communication is pivotal for addressing complex collaborative tasks in multi-agent reinforcement learning (MARL). Yet, limited communication bandwidth and dynamic, intricate environmental topologies present significant challenges in identifying high-value communication partners. Agents must consequently select collaborators under uncertainty, lacking a priori knowledge of which partners can deliver task-critical information. To this end, we propose Interference-Aware K-Step Reachable Communication (IA-KRC), a novel framework that enhances cooperation via two core components: (1) a K-Step reachability protocol that confines message passing to physically accessible neighbors, and (2) an interference-prediction module that optimizes partner choice by minimizing interference while maximizing utility. Compared to existing methods, IA-KRC enables substantially more persistent and efficient cooperation despite environmental interference. Comprehensive evaluations confirm that IA-KRC achieves superior performance compared to state-of-the-art baselines, while demonstrating enhanced robustness and scalability in complex topological and highly dynamic multi-agent scenarios.