mlsanityData to 14 Oct 2026

Reinforcement Learning Inspired Black-box Adversarial Attacks for Computer Vision

arXiv 2609.24249 ↗First posted 21 Sep 2026cs.LG cs.CR cs.CVcited by 0

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

Abstract (arXiv)

Neural networks, both convolution or transformer based, are essential for modern computer vision systems. However, they are vulnerable to small perturbations, almost imperceptible to humans, which significantly alter the model's prediction. These adversarial attacks are often considered to be a significant threat to the implementation of neural networks in safety-critical applications. Most attacks utilize the white-box threat model and therefore require full access to the target model, making them unrealistic to use in practice. We propose a novel approach under the more realistic black-box threat model that utilizes concepts from reinforcement learning to optimize perturbations with a non-differentiable target model. Reinforcement learning algorithms have already been optimized to be query efficient, making them an ideal starting point when designing black-box adversarial attacks. We show the success of our reinforcement learning inspired black-box adversarial attack (RIBA) in generating adversarial perturbations using only a small number of queries to the target model, by comparing it to state of the art attacks on different models on the Cifar10 and ImageNet data sets. RIBA takes 25.4% fewer median queries to generate attacked images against a ResNet-18 on Cifar10 and 22.5% fewer median queries to fool a Vit-B/16 model on ImageNet. Additionally, we demonstrate that RIBA can match the performance of white-box attacks on an adversarially trained model.

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 31 references that are in the map. First posted on arXiv 21 Sep 2026.

Builds on

23 of 31 references.

References · 23

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  6. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  7. 2013Intriguing properties of neural networkscited by 5,155
  8. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  9. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  10. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
  11. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 685
  12. 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 666
11 more
  1. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 661
  2. 2018Black-box Adversarial Attacks with Limited Queries and Informationcited by 501
  3. 2019Simple Black-box Adversarial Attackscited by 251
  4. 2018Prior Convictions: Black-Box Adversarial Attacks with Bandits and Priorscited by 146
  5. 2021Stochastic Variance Reduced Ensemble Adversarial Attack for Boosting the Adversarial Transferabilitycited by 84
  6. 2019ZO-AdaMM: Zeroth-Order Adaptive Momentum Method for Black-Box Optimizationcited by 76
  7. 2024Adversarial Robustness Limits via Scaling-Law and Human-Alignment Studiescited by 38
  8. 2020Learning Black-Box Attackers with Transferable Priors and Query Feedbackcited by 32
  9. 2020Towards Query-Efficient Black-Box Adversary with Zeroth-Order Natural Gradient Descentcited by 18
  10. 2020Adversarial Eigen Attack on Black-Box Modelscited by 6
  11. 2022Universal Distributional Decision-based Black-box Adversarial Attack with Reinforcement Learningcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Reinforcement Learning Platform for Adversarial Black-box Attacks with Custom Distortion Filters
  2. 2025SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning
  3. 2025Adversarial Agents: Black-Box Evasion Attacks with Reinforcement Learning
  4. 2024Theoretical Corrections and the Leveraging of Reinforcement Learning to Enhance Triangle Attack
  5. 2026Boosting Transferable Adversarial Attacks against Deep Reinforcement Learning
  6. 2026Contract And Conquer: How to Provably Compute Adversarial Examples for a Black-Box Model?
  7. 2024Efficient Black-box Adversarial Attacks via Bayesian Optimization Guided by a Function Prior
  8. 2025Amnesia as a Catalyst for Enhancing Black Box Pixel Attacks in Image Classification and Object Detection