mlsanityData to 14 Oct 2026

SAJA: A State-Action Joint Attack Framework on Multi-Agent Deep Reinforcement Learning

arXiv 2510.13262 ↗First posted 15 Oct 2025cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Multi-Agent Deep Reinforcement Learning (MADRL) has shown potential for cooperative and competitive tasks such as autonomous driving and strategic gaming. However, models trained by MADRL are vulnerable to adversarial perturbations on states and actions. Therefore, it is essential to investigate the robustness of MADRL models from an attack perspective. Existing studies focus on either state-only attacks or action-only attacks, but do not consider how to effectively joint them. Simply combining state and action perturbations such as randomly perturbing states and actions does not exploit their potential synergistic effects. In this paper, we propose the State-Action Joint Attack (SAJA) framework that has a good synergistic effects. SAJA consists of two important phases: (1) In the state attack phase, a multi-step gradient ascent method utilizes both the actor network and the critic network to compute an adversarial state, and (2) in the action attack phase, based on the perturbed state, a second gradient ascent uses the critic network to craft the final adversarial action. Additionally, a heuristic regularizer measuring the distance between the perturbed actions and the original clean ones is added into the loss function to enhance the effectiveness of the critic's guidance. We evaluate SAJA in the Multi-Agent Particle Environment (MPE), demonstrating that (1) it outperforms and is more stealthy than state-only or action-only attacks, and (2) existing state or action defense methods cannot defend its attacks.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 35 references that are in the map. First posted on arXiv 15 Oct 2025.

Builds on

20 of 35 references.

References · 20

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,558
  4. 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,103
  5. 2021The Surprising Effectiveness of PPO in Cooperative, Multi-Agent Gamescited by 942
  6. 2017Counterfactual Multi-Agent Policy Gradientscited by 751
  7. 2019The StarCraft Multi-Agent Challengecited by 597
  8. 2017Adversarial Attacks on Neural Network Policiescited by 320
  9. 2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 238
  10. 2013The Complexity of Decentralized Control of Markov Decision Processescited by 231
  11. 2019Adversarial Policies: Attacking Deep Reinforcement Learningcited by 214
  12. 2020FACMAC: Factored Multi-Agent Centralised Policy Gradientscited by 178
8 more
  1. 2019Action Robust Reinforcement Learning and Applications in Continuous Controlcited by 174
  2. 2021Robust Reinforcement Learning on State Observations with Learned Optimal Adversarycited by 135
  3. 2021Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RLcited by 55
  4. 2021Adversarial Attacks On Multi-Agent Communicationcited by 49
  5. 2020On the Robustness of Cooperative Multi-Agent Reinforcement Learningcited by 39
  6. 2021ROMAX: Certifiably Robust Deep Multiagent Reinforcement Learning via Convex Relaxationcited by 17
  7. 2023Attacking Cooperative Multi-Agent Reinforcement Learning by Adversarial Minority Influencecited by 17
  8. 2023Robustness Testing for Multi-Agent Reinforcement Learning: State Perturbations on Critical Agentscited by 7

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. 2025Neutral Agent-based Adversarial Policy Learning against Deep Reinforcement Learning in Multi-party Open Systems
  3. 2024SUB-PLAY: Adversarial Policies against Partially Observed Multi-Agent Reinforcement Learning Systems
  4. 2025Enhancing Cooperative Multi-Agent Reinforcement Learning with State Modelling and Adversarial Exploration
  5. 2024Multi-Agent Actor-Critics in Autonomous Cyber Defense
  6. 2025Policy Disruption in Reinforcement Learning:Adversarial Attack with Large Language Models and Critical State Identification
  7. 2024Robust Deep Reinforcement Learning Through Adversarial Attacks and Training : A Survey
  8. 2025Robust and Diverse Multi-Agent Learning via Rational Policy Gradient