mlsanityData to 14 Oct 2026

Boosting Transferable Adversarial Attacks against Deep Reinforcement Learning

arXiv 2610.06083 ↗First posted 5 Oct 2026cs.LG cs.AIcited by 0

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

Abstract (arXiv)

Most adversarial attacks on deep reinforcement learning (DRL) assume white-box access to the victim policy, which rarely holds in practice. This paper studies transfer-based black-box attacks on DRL: the attacker crafts observation perturbations on a white-box surrogate agent and feeds them to an unknown victim. We formulate the attack as return minimization under a per-step perturbation budget. We first show that transplanting transferable image-classification attacks (FGSM, MI-FGSM, and NI-FGSM) with a per-step objective yields perturbations that transfer but are no stronger than random noise of the same budget. We then propose a trajectory-level attack that optimizes a sequence of perturbations over a receding horizon through a differentiable model of the environment and a temperature-smoothed surrogate policy, with the same optimizers. On CartPole-v1 with ten DQN and DDQN agents and 100 surrogate--victim pairs, the trajectory-level attack outperforms per-step attacks and random noise in the white-box, cross-model, and cross-algorithm settings.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Adversarial robustness and RL in finance, one of 15 directions in Deep reinforcement learning methods. Builds on 23 of its 32 references that are in the map. First posted on arXiv 5 Oct 2026.

Builds on

23 of 32 references.

Closest ancestors · 1

Same direction, newest first

  1. 2026PIMbot: A Self-Adaptive Attack Framework for Adversarial Manipulation of Multi-Robot Reinforcement Learning21 May

Other references · 22

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2016Adversarial examples in the physical worldcited by 1,769
  4. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,745
  5. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 599
  6. 2019Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attackscited by 355
  7. 2017Adversarial Attacks on Neural Network Policiescited by 320
  8. 2021Enhancing the Transferability of Adversarial Attacks through Variance Tuningcited by 244
  9. 2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 238
  10. 2017Tactics of Adversarial Attack on Deep Reinforcement Learning Agentscited by 166
  11. 2017Self-supervised Deep Reinforcement Learning with Generalized Computation Graphs for Robot Navigationcited by 77
  12. 2021Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RLcited by 55
10 more
  1. 2023Sibling-Attack: Rethinking Transferable Adversarial Attacks against Face Recognitioncited by 34
  2. 2023Adversarial Robust Deep Reinforcement Learning Requires Redefining Robustnesscited by 19
  3. 2023White-Box Multi-Objective Adversarial Attack on Dialogue Generationcited by 14
  4. 2023Dynamic Transformers Provide a False Sense of Efficiencycited by 14
  5. 2021Deep Reinforcement Learning Policies Learn Shared Adversarial Features Across MDPscited by 12
  6. 2023Detecting Adversarial Directions in Deep Reinforcement Learning to Make Robust Decisionscited by 7
  7. 2023PIMbot: Policy and Incentive Manipulation for Multi-Robot Reinforcement Learning in Social Dilemmascited by 6
  8. 2024Understanding and Diagnosing Deep Reinforcement Learningcited by 6
  9. 2024Transferable Adversarial Attacks against ASRcited by 4
  10. 2026SPEAR: Distilling Domain-Adaptive Reasoning Skeletons via Sequential Symbolic Alignment in Reinforcement Learningcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025State-Aware Perturbation Optimization for Robust Deep Reinforcement Learning
  2. 2026Reinforcement Learning Inspired Black-box Adversarial Attacks for Computer Vision
  3. 2025Adversarial Agents: Black-Box Evasion Attacks with Reinforcement Learning
  4. 2025Diffusion Guided Adversarial State Perturbations in Reinforcement Learning
  5. 2025Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks
  6. 2025Rethinking Adversarial Attacks in Reinforcement Learning from Policy Distribution Perspective
  7. 2025Enhancing Security in Deep Reinforcement Learning: A Comprehensive Survey on Adversarial Attacks and Defenses
  8. 2025Towards Robust Deep Reinforcement Learning against Environmental State Perturbation