mlsanityData to 14 Oct 2026

Distributed Detection of Adversarial Attacks in Multi-Agent Reinforcement Learning with Continuous Action Space

arXiv 2508.15764 ↗First posted 21 Aug 2025cs.LG cs.MAcited by 0

Not on the map. Why?

Abstract (arXiv)

We address the problem of detecting adversarial attacks against cooperative multi-agent reinforcement learning with continuous action space. We propose a decentralized detector that relies solely on the local observations of the agents and makes use of a statistical characterization of the normal behavior of observable agents. The proposed detector utilizes deep neural networks to approximate the normal behavior of agents as parametric multivariate Gaussian distributions. Based on the predicted density functions, we define a normality score and provide a characterization of its mean and variance. This characterization allows us to employ a two-sided CUSUM procedure for detecting deviations of the normality score from its mean, serving as a detector of anomalous behavior in real-time. We evaluate our scheme on various multi-agent PettingZoo benchmarks against different state-of-the-art attack methods, and our results demonstrate the effectiveness of our method in detecting impactful adversarial attacks. Particularly, it outperforms the discrete counterpart by achieving AUC-ROC scores of over 0.95 against the most impactful attacks in all evaluated environments.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 25 of its 46 references that are in the map. First posted on arXiv 21 Aug 2025.

Builds on

25 of 46 references.

References · 25

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,558
  4. 2018Distributed Prioritized Experience Replaycited by 269
  5. 2016LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detectioncited by 215
  6. 2019Adversarial Policies: Attacking Deep Reinforcement Learningcited by 214
  7. 2017Robust Deep Reinforcement Learning with Adversarial Attackscited by 154
  8. 2017Vulnerability of Deep Reinforcement Learning to Policy Induction Attackscited by 91
  9. 2021Scaling Multi-Agent Reinforcement Learning with Selective Parameter Sharingcited by 87
  10. 2020On the Robustness of Cooperative Multi-Agent Reinforcement Learningcited by 39
  11. 2023Robust Multi-Agent Reinforcement Learning via Adversarial Regularization: Theoretical Foundation and Stable Algorithmscited by 25
  12. 2021Continuous Control with Action Quantization from Demonstrationscited by 22
13 more
  1. 2021Recurrent Off-policy Baselines for Memory-based Continuous Controlcited by 21
  2. 2022Towards Comprehensive Testing on the Robustness of Cooperative Multi-agent Reinforcement Learningcited by 21
  3. 2020Sequential Anomaly Detection using Inverse Reinforcement Learningcited by 20
  4. 2020Multi-Scale One-Class Recurrent Neural Networks for Discrete Event Sequence Anomaly Detectioncited by 19
  5. 2017Detecting Adversarial Attacks on Neural Network Policies with Visual Foresightcited by 17
  6. 2021Adversarial attacks in consensus-based multi-agent reinforcement learningcited by 17
  7. 2020Policy Entropy for Out-of-Distribution Classificationcited by 11
  8. 2021Real-time Adversarial Perturbations against Deep Reinforcement Learning Policies: Attacks and Defensescited by 9
  9. 2021A Distance-based Anomaly Detection Framework for Deep Reinforcement Learningcited by 5
  10. 2023Robust Communicative Multi-Agent Reinforcement Learning with Active Defensecited by 5
  11. 2024Discretizing Continuous Action Space with Unimodal Probability Distributions for On-Policy Reinforcement Learningcited by 4
  12. 2024OIL-AD: An Anomaly Detection Framework for Sequential Decision Sequencescited by 3
  13. 2025Quickest Detection of Adversarial Attacks Against Correlated Equilibriacited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Constrained Black-Box Attacks Against Cooperative Multi-Agent Reinforcement Learning
  2. 2025Adversarial Agents: Black-Box Evasion Attacks with Reinforcement Learning
  3. 2024CuDA2: An approach for Incorporating Traitor Agents into Cooperative Multi-Agent Systems
  4. 2026Interaction-Breaking Adversarial Learning Framework for Robust Multi-Agent Reinforcement Learning
  5. 2025Neutral Agent-based Adversarial Policy Learning against Deep Reinforcement Learning in Multi-party Open Systems
  6. 2024SUB-PLAY: Adversarial Policies against Partially Observed Multi-Agent Reinforcement Learning Systems
  7. 2025SAJA: A State-Action Joint Attack Framework on Multi-Agent Deep Reinforcement Learning
  8. 2026Robust Multi-Agent Path Finding under Observation Attacks: A Principled Adversarial-Plus-Smoothing Training Recipe