Interaction-Breaking Adversarial Learning Framework for Robust Multi-Agent Reinforcement Learning

TL;DR

This paper introduces an adversarial framework that enhances the robustness of multi-agent reinforcement learning by disrupting interactions, leading to more reliable coordination under perturbations.

Contribution

It proposes an interaction-breaking adversarial learning framework that specifically targets and disrupts agent interactions, improving robustness beyond existing methods.

Findings

Improves robustness over existing MARL baselines under various attack scenarios.

Achieves stronger performance even when some agents are missing.

Uses an information-theoretic approach to construct effective attacks.

Abstract

Cooperation is central to multi-agent reinforcement learning (MARL), yet learned coordination can be fragile when external perturbations disrupt inter-agent interactions. Prior robust MARL methods have primarily considered value-oriented attacks, leaving a gap in robustness when interaction structures themselves are corrupted. In this paper, we propose an interaction-breaking adversarial learning (IBAL) framework that takes an information-theoretic view to construct attacks that impede coordination by perturbing agents' observations and actions, and trains agents to perform reliably under such disruptions. Empirically, our approach improves robustness over existing robust MARL baselines across diverse attack settings and yields stronger performance even under agent-missing scenarios.