Diversified Adversarial Attacks based on Conjugate Gradient Method

Keiichiro Yamamura, Haruki Sato, Nariaki Tateiwa, Nozomi Hata, Toru Mitsutake, Issa Oe, Hiroki Ishikura, Katsuki Fujisawa

Abstract: Deep learning models are vulnerable to adversarial examples, and adversarial attacks used to generate such examples have attracted considerable research interest. Although existing methods based on...