mlsanityData to 14 Oct 2026

Robust Deep Reinforcement Learning against Adversarial Behavior Manipulation

arXiv 2406.03862 ↗First posted 6 Jun 2024cs.LG cs.AIcited by 2

On the map: Deep reinforcement learning methods → Adversarial robustness and RL in finance

Abstract (arXiv)

This study investigates behavior-targeted attacks on reinforcement learning and their countermeasures. Behavior-targeted attacks aim to manipulate the victim's behavior as desired by the adversary through adversarial interventions in state observations. Existing behavior-targeted attacks have some limitations, such as requiring white-box access to the victim's policy. To address this, we propose a novel attack method using imitation learning from adversarial demonstrations, which works under limited access to the victim's policy and is environment-agnostic. In addition, our theoretical analysis proves that the policy's sensitivity to state changes impacts defense performance, particularly in the early stages of the trajectory. Based on this insight, we propose time-discounted regularization, which enhances robustness against attacks while maintaining task performance. To the best of our knowledge, this is the first defense strategy specifically designed for behavior-targeted attacks.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Adversarial robustness and RL in finance, one of 15 directions in Deep reinforcement learning methods. Builds on 55 of its 76 references that are in the map. First posted on arXiv 6 Jun 2024.

Builds on

55 of 76 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024SleeperNets: Universal Backdoor Poisoning Attacks Against Reinforcement Learning Agents30 May
  2. 2024Belief-Enriched Pessimistic Q-Learning against Adversarial State Perturbations6 Mar
  3. 2024Robust Deep Reinforcement Learning Through Adversarial Attacks and Training : A Survey1 Mar
  4. 2024Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies20 Feb
  5. 2024Towards Optimal Adversarial Robust Q-learning with Bellman Infinity-error3 Feb

Other references · 50

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  4. 2016Generative Adversarial Imitation Learningcited by 1,597
  5. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
  6. 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,103
  7. 2020Deep Reinforcement Learning for Autonomous Driving: A Surveycited by 721
  8. 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 567
  9. 2018Efficient Neural Network Robustness Certification with General Activation Functionscited by 445
  10. 2017Adversarial Attacks on Neural Network Policiescited by 320
  11. 2019Reinforcement Learning in Healthcare: A Surveycited by 310
  12. 2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 238
38 more
  1. 2017Deep Reinforcement Learning framework for Autonomous Drivingcited by 225
  2. 2019Adversarial Policies: Attacking Deep Reinforcement Learningcited by 214
  3. 2019Towards Stable and Efficient Training of Verifiably Robust Neural Networkscited by 205
  4. 2018Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learningcited by 204
  5. 2018Generative Adversarial Imitation from Observationcited by 184
  6. 2017Tactics of Adversarial Attack on Deep Reinforcement Learning Agentscited by 166
  7. 2017Robust Deep Reinforcement Learning with Adversarial Attackscited by 154
  8. 2021Robust Reinforcement Learning on State Observations with Learned Optimal Adversarycited by 135
  9. 2019A Convex Relaxation Barrier to Tight Robustness Verification of Neural Networkscited by 125
  10. 2020Policy Teaching via Environment Poisoning: Training-time Adversarial Attacks against Reinforcement Learningcited by 81
  11. 2020Robust Deep Reinforcement Learning through Adversarial Losscited by 75
  12. 2023Data Quality in Imitation Learningcited by 75
  13. 2021Off-Policy Imitation Learning from Observationscited by 70
  14. 2020Stealthy and Efficient Adversarial Attacks against Deep Reinforcement Learningcited by 63
  15. 2021Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RLcited by 55
  16. 2022Efficient Adversarial Training without Attacking: Worst-Case-Aware Robust Reinforcement Learningcited by 43
  17. 2021CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothingcited by 42
  18. 2021Policy Smoothing for Provably Robust Reinforcement Learningcited by 40
  19. 2020Vulnerability-Aware Poisoning Mechanism for Online RL with Unknown Dynamicscited by 38
  20. 2019Attacking Vision-based Perception in End-to-End Autonomous Driving Modelscited by 31
  21. 2023BadRL: Sparse Targeted Backdoor Attack Against Reinforcement Learningcited by 30
  22. 2023Robust Multi-Agent Reinforcement Learning via Adversarial Regularization: Theoretical Foundation and Stable Algorithmscited by 25
  23. 2022Understanding the Limits of Poisoning Attacks in Episodic Reinforcement Learningcited by 21
  24. 2023Adversarial Robust Deep Reinforcement Learning Requires Redefining Robustnesscited by 19
  25. 2018Sequential Attacks on Agents for Long-Term Adversarial Goalscited by 16
  26. 2021Investigating Vulnerabilities of Deep Neural Policiescited by 15
  27. 2021Understanding Adversarial Attacks on Observations in Deep Reinforcement Learningcited by 13
  28. 2021Targeted Attack on Deep RL-based Autonomous Driving with Learned Visual Patternscited by 13
  29. 2017Regret Minimization for Partially Observable Deep Reinforcement Learningcited by 12
  30. 2019CopyCAT: Taking Control of Neural Policies with Constant Attackscited by 12
  31. 2023Optimal Attack and Defense for Reinforcement Learningcited by 12
  32. 2020Minimax Regret Optimisation for Robust Planning in Uncertain Markov Decision Processescited by 10
  33. 2022Efficient Reward Poisoning Attacks on Online Deep Reinforcement Learningcited by 9
  34. 2023Reward Certification for Policy Smoothed Reinforcement Learningcited by 7
  35. 2022Consistent Attack: Universal Adversarial Perturbation on Embodied Vision Navigationcited by 6
  36. 2023Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learningcited by 5
  37. 2024Adversarial Imitation Learning via Boostingcited by 5
  38. 2023Regret-Based Defense in Adversarial Reinforcement Learningcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Demonstration-Guided Observation Attacks on Black-Box Safe Reinforcement Learning Controllers for Robotic Systemscited by 0
  2. 2026Advantage-based Temporal Attack in Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024RAT: Adversarial Attacks on Deep Reinforcement Agents for Targeted Behaviors
  2. 2025Policy Disruption in Reinforcement Learning:Adversarial Attack with Large Language Models and Critical State Identification
  3. 2022Imitating Opponent to Win: Adversarial Policy Imitation Learning in Two-player Competitive Games
  4. 2025How Vulnerable Is My Learned Policy? Universal Adversarial Perturbation Attacks On Modern Behavior Cloning Policies
  5. 2025Towards Robust Deep Reinforcement Learning against Environmental State Perturbation
  6. 2024Towards Robust Policy: Enhancing Offline Reinforcement Learning with Adversarial Attacks and Defenses
  7. 2026BehaviorGuard: Online Backdoor Defense for Deep Reinforcement Learning
  8. 2025State-Aware Perturbation Optimization for Robust Deep Reinforcement Learning