Threat Model-Agnostic Adversarial Defense using Diffusion Models

Understand

Deep Neural Networks (DNNs) are highly sensitive to imperceptible malicious perturbations, known as adversarial attacks.

• Following the discovery of this vulnerability in real-world imaging and vision applications, the associated safety concerns have attracted vast research attention, and many defense techniques have been developed.
• Most of these defense methods rely on adversarial training (AT) -- training the classification network on images perturbed according to a specific threat model, which defines the magnitude of the allowed modification.
• Although AT leads to promising results, training on a specific threat model fails to generalize to other types of perturbations.