Robustness via curvature regularization, and vice versa
CVPR, Volume abs/1811.09716, 2019.
State-of-the-art classifiers have been shown to be largely vulnerable to adversarial perturbations. One of the most effective strategies to improve robustness is adversarial training. In this paper, we investigate the effect of adversarial training on the geometry of the classification landscape and decision boundaries. We show in particu...More
PPT (Upload PPT)