Adversarial Training Generalizes Data-dependent Spectral Norm Regularization

06/04/2019
by   Kevin Roth, et al.
12

We establish a theoretical link between adversarial training and operator norm regularization for deep neural networks. Specifically, we show that adversarial training is a data-dependent generalization of spectral norm regularization. This intriguing connection provides fundamental insights into the origin of adversarial vulnerability and hints at novel ways to robustify and defend against adversarial attacks. We provide extensive empirical evidence to support our theoretical results.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset