Guided Diffusion Model for Adversarial Purification. (arXiv:2205.14969v3 [cs.CV] UPDATED)

With wider application of deep neural networks (DNNs) in various algorithms
and frameworks, security threats have become one of the concerns. Adversarial
attacks disturb DNN-based image classifiers, in which attackers can
intentionally add imperceptible adversarial perturbations on input images to
fool the classifiers. In this paper, we propose a novel purification approach,
referred to as guided diffusion model for purification (GDMP), to help protect
classifiers from adversarial attacks. The core of our approach is to embed
purification into the diffusion …

arxiv cv diffusion diffusion model