On the Structural Sensitivity of Deep Convolutional Networks to the Directions of Fourier Basis Functions

Yusuke Tsuzuku, Issei Sato; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 51-60

Abstract


Data-agnostic quasi-imperceptible perturbations on inputs are known to degrade recognition accuracy of deep convolutional networks severely. This phenomenon is considered to be a potential security issue. Moreover, some results on statistical generalization guarantees indicate that the phenomena can be a key to improve the networks' generalization. However, the characteristics of the shared directions of such harmful perturbations remain unknown. Our primal finding is that convolutional networks are sensitive to the directions of Fourier basis functions. We derived the property by specializing a hypothesis of the cause of the sensitivity, known as the linearity of neural networks, to convolutional networks and empirically validated it. As a byproduct of the analysis, we propose an algorithm to create shift-invariant universal adversarial perturbations available in black-box settings.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Tsuzuku_2019_CVPR,
author = {Tsuzuku, Yusuke and Sato, Issei},
title = {On the Structural Sensitivity of Deep Convolutional Networks to the Directions of Fourier Basis Functions},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}