Reinforcement Learning Platform for Adversarial Black-box Attacks with Custom Distortion Filters
Soumyendu Sarkar, Ashwin Ramesh Babu, Sajad Mousavi, Vineet Gundecha, Sahand Ghorbanpour, Avisek Naug, Ricardo Luna Gutiérrez, Antonio Guillen
Abstract
We present a Reinforcement Learning Platform for Adversarial Black-box untargeted and targeted attacks, RLAB, that allows users to select from various distortion filters to create adversarial examples. The platform uses a Reinforcement Learning agent to add minimum distortion to input images while still causing misclassification by the target model. The agent uses a novel dual-action method to explore the input image at each step to identify sensitive regions for adding distortions while removing noises that have less impact on the target model. This dual action leads to faster and more efficient convergence of the attack. The platform can also be used to measure the robustness of image classification models against specific distortion types. Also, retraining the model with adversarial samples significantly improved robustness when evaluated on benchmark datasets. The proposed platform outperforms state-of-the-art methods in terms of the average number of queries required to cause misclassification. This advances trustworthiness with a positive social impact.
BibTeX
@article{Sarkar_Ramesh Babu_Mousavi_Gundecha_Ghorbanpour_Naug_Luna Gutiérrez_Guillen_Rengarajan_2025, title={Reinforcement Learning Platform for Adversarial Black-box Attacks with Custom Distortion Filters}, volume={39}, url={https://ojs.aaai.org/index.php/AAAI/article/view/34976}, DOI={10.1609/aaai.v39i26.34976}, abstractNote={We present a Reinforcement Learning Platform for Adversarial Black-box untargeted and targeted attacks, RLAB, that allows users to select from various distortion filters to create adversarial examples. The platform uses a Reinforcement Learning agent to add minimum distortion to input images while still causing misclassification by the target model. The agent uses a novel dual-action method to explore the input image at each step to identify sensitive regions for adding distortions while removing noises that have less impact on the target model. This dual action leads to faster and more efficient convergence of the attack. The platform can also be used to measure the robustness of image classification models against specific distortion types. Also, retraining the model with adversarial samples significantly improved robustness when evaluated on benchmark datasets. The proposed platform outperforms state-of-the-art methods in terms of the average number of queries required to cause misclassification. This advances trustworthiness with a positive social impact.}, number={26}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Sarkar, Soumyendu and Ramesh Babu, Ashwin and Mousavi, Sajad and Gundecha, Vineet and Ghorbanpour, Sahand and Naug, Avisek and Luna Gutiérrez, Ricardo and Guillen, Antonio and Rengarajan, Desik}, year={2025}, month={Apr.}, pages={27628-27635} }