Robustness and Visual Explanation for Black Box Image, Video, and ECG Signal Classification with Reinforcement Learning

arXiv:2403.18985v1 Announce Type: new
Abstract: We present a generic Reinforcement Learning (RL) framework optimized for crafting adversarial attacks on different model types spanning from ECG signal analysis (1D), image classification (2D), and video classification (3D). The framework focuses on identifying sensitive regions and inducing misclassifications with minimal distortions and various distortion types. The novel RL method outperforms state-of-the-art methods for all three applications, proving its efficiency. Our RL approach produces superior localization masks, enhancing interpretability for image classification and ECG …

abstract adversarial adversarial attacks analysis arxiv attacks black box box classification cs.ai cs.cr cs.cv cs.lg cs.ma framework image reinforcement reinforcement learning robustness signal type types video video classification visual