AdvDiff: Generating Unrestricted Adversarial Examples using Diffusion Models

arXiv:2307.12499v3 Announce Type: replace
Abstract: Unrestricted adversarial attacks present a serious threat to deep learning models and adversarial defense techniques. They pose severe security problems for deep learning applications because they can effectively bypass defense mechanisms. However, previous attack methods often utilize Generative Adversarial Networks (GANs), which are not theoretically provable and thus generate unrealistic examples by incorporating adversarial objectives, especially for large-scale datasets like ImageNet. In this paper, we propose a new method, called AdvDiff, to generate unrestricted adversarial …

abstract adversarial adversarial attacks adversarial examples applications arxiv attack methods attacks cs.cv cs.lg deep learning defense diffusion diffusion models examples gans generative generative adversarial networks networks security threat type