Are Perceptually-Aligned Gradients a General Property of Robust Classifiers?
Abstract
For a standard convolutional neural network, optimizing over the input pixels to maximize the score of some target class will generally produce a grainy-looking version of the original image. However, Santurkar et al. (2019) demonstrated that for adversarially-trained neural networks, this optimization produces images that uncannily resemble the target class. In this paper, we show that these "perceptually-aligned gradients" also occur under randomized smoothing, an alternative means of constructing adversarially-robust classifiers. Our finding supports the hypothesis that perceptually-aligned gradients may be a general property of robust classifiers. We hope that our results will inspire research aimed at explaining this link between perceptually-aligned gradients and adversarial robustness.
Cite
@article{arxiv.1910.08640,
title = {Are Perceptually-Aligned Gradients a General Property of Robust Classifiers?},
author = {Simran Kaur and Jeremy Cohen and Zachary C. Lipton},
journal= {arXiv preprint arXiv:1910.08640},
year = {2019}
}
Comments
To appear in the "Science Meets Engineering of Deep Learning" Workshop at NeurIPS 2019