mlsanityData to 14 Oct 2026

Certifiably Robust Policy Learning against Adversarial Communication in Multi-agent Systems

arXiv 2206.10158 ↗First posted 21 Jun 2022cs.LG cs.MAcited by 7

Not on the map. Why?

Abstract (arXiv)

Communication is important in many multi-agent reinforcement learning (MARL) problems for agents to share information and make good decisions. However, when deploying trained communicative agents in a real-world application where noise and potential attackers exist, the safety of communication-based policies becomes a severe issue that is underexplored. Specifically, if communication messages are manipulated by malicious attackers, agents relying on untrustworthy communication may take unsafe actions that lead to catastrophic consequences. Therefore, it is crucial to ensure that agents will not be misled by corrupted communication, while still benefiting from benign communication. In this work, we consider an environment with N agents, where the attacker may arbitrarily change the communication from any C<fracN-12 agents to a victim agent. For this strong threat model, we propose a certifiable defense by constructing a message-ensemble policy that aggregates multiple randomly ablated message sets. Theoretical analysis shows that this message-ensemble policy can utilize benign communication while being certifiably robust to adversarial communication, regardless of the attacking algorithm. Experiments in multiple environments verify that our defense significantly improves the robustness of trained policies against various types of attacks.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 37 of its 50 references that are in the map. First posted on arXiv 21 Jun 2022.

Builds on

37 of 50 references.

References · 37

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  4. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
  5. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  6. 2018Robustness May Be at Odds with Accuracycited by 880
  7. 2018QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learningcited by 708
  8. 2016Learning to Communicate with Deep Multi-Agent Reinforcement Learningcited by 575
  9. 2017Robust Adversarial Reinforcement Learningcited by 451
  10. 2016Learning Multiagent Communication with Backpropagationcited by 366
  11. 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 328
  12. 2017Adversarial Attacks on Neural Network Policiescited by 320
25 more
  1. 2016Safe, Multi-Agent, Reinforcement Learning for Autonomous Drivingcited by 311
  2. 2018Certified Defenses against Adversarial Examplescited by 306
  3. 2018Adversarial Attacks and Defences: A Surveycited by 259
  4. 2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 238
  5. 2017Deep Reinforcement Learning framework for Autonomous Drivingcited by 225
  6. 2020PettingZoo: Gym for Multi-Agent Reinforcement Learningcited by 220
  7. 2018Model-Ensemble Trust-Region Policy Optimizationcited by 219
  8. 2019Adversarial Policies: Attacking Deep Reinforcement Learningcited by 214
  9. 2018TarMAC: Targeted Multi-Agent Communicationcited by 208
  10. 2019Action Robust Reinforcement Learning and Applications in Continuous Controlcited by 174
  11. 2021Robust Reinforcement Learning on State Observations with Learned Optimal Adversarycited by 135
  12. 2020Who2com: Collaborative Perception via Learnable Handshake Communicationcited by 127
  13. 2020Robust Deep Reinforcement Learning through Adversarial Losscited by 75
  14. 2019Robustness Certificates for Sparse Adversarial Attacks by Randomized Ablationcited by 60
  15. 2021Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RLcited by 55
  16. 2021Adversarial Attacks On Multi-Agent Communicationcited by 49
  17. 2019Certified Adversarial Robustness for Deep Reinforcement Learningcited by 47
  18. 2021CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothingcited by 42
  19. 2021Policy Smoothing for Provably Robust Reinforcement Learningcited by 40
  20. 2020The Emergence of Adversarial Communication in Multi-Agent Reinforcement Learningcited by 28
  21. 2019Online Robustness Training for Deep Reinforcement Learningcited by 26
  22. 2021Mis-spoke or mis-lead: Achieving Robustness in Multi-Agent Communicative Reinforcement Learningcited by 18
  23. 2020Gaussian Process Based Message Filtering for Robust Multi-Agent Cooperation in the Presence of Adversarial Communicationcited by 11
  24. 2019Multi-Agent Image Classification via Reinforcement Learningcited by 8
  25. 2014Off-Policy Shaping Ensembles in Reinforcement Learningcited by 2

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 4; 2024: 0; 2025: 1.

7 papers.

  1. 2022What is the Solution for State-Adversarial Multi-Agent Reinforcement Learning?cited by 22
  2. 2023Instructed Diffuser with Temporal Condition Guidance for Offline Reinforcement Learningcited by 13
  3. 2022Certified Policy Smoothing for Cooperative Multi-Agent Reinforcement Learningcited by 7
  4. 2023Rethinking Adversarial Policies: A Generalized Attack Formulation and Provable Defense in RLcited by 7
  5. 2023Robust Multi-agent Communication via Multi-view Message Certificationcited by 3
  6. 2025RAGPart & RAGMask: Retrieval-Stage Defenses Against Corpus Poisoning in Retrieval-Augmented Generationcited by 3
  7. 2023Optimal Cost Constrained Adversarial Attacks For Multiple Agent Systemscited by 1