Reinforcement Learning Platform for Adversarial Black-box Attacks with Custom Distortion Filters
This work addresses the need for more efficient and customizable adversarial attacks to evaluate and improve model robustness in machine learning security.
The paper tackles the problem of generating adversarial examples for black-box attacks on image classification models by introducing a reinforcement learning platform (RLAB) that uses a novel dual-action method to add minimal distortion, resulting in fewer queries needed for misclassification compared to state-of-the-art methods.
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.