Robust Overfitting Does Matter: Test-Time Adversarial Purification With FGSM

arXiv:2403.11448v1 Announce Type: new
Abstract: Numerous studies have demonstrated the susceptibility of deep neural networks (DNNs) to subtle adversarial perturbations, prompting the development of many advanced adversarial defense methods aimed at mitigating adversarial attacks. Current defense strategies usually train DNNs for a specific adversarial attack method and can achieve good robustness in defense against this type of adversarial attack. Nevertheless, when subjected to evaluations involving unfamiliar attack modalities, empirical evidence reveals a pronounced deterioration in the robustness of DNNs. Meanwhile, …

abstract advanced adversarial adversarial attacks arxiv attacks cs.cv current defense development good matter networks neural networks overfitting prompting robust robustness strategies studies test train type