Improving Global Adversarial Robustness Generalization With Adversarially Trained GAN

no code implementations • 8 Mar 2021 • DeSheng Wang, Weidong Jin, Yunpu Wu, Aamir Khan

In order to defend against the adversarial perturbations, adversarially trained GAN (ATGAN) is proposed to improve the adversarial robustness generalization of the state-of-the-art CNNs trained by adversarial training.

Adversarial Defense Adversarial Robustness +2