High-Frequency Component Helps Explain the Generalization of Convolutional Neural Networks

Haohan Wang, Xindi Wu, Zeyi Huang, Eric P. Xing; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 8684-8694

Abstract


We investigate the relationship between the frequency spectrum of image data and the generalization behavior of convolutional neural networks (CNN). We first notice CNN's ability in capturing the high-frequency components of images. These high-frequency components are almost imperceptible to a human. Thus the observation leads to multiple hypotheses that are related to the generalization behaviors of CNN, including a potential explanation for adversarial examples, a discussion of CNN's trade-off between robustness and accuracy, and some evidence in understanding training heuristics.

Related Material


[pdf] [supp] [arXiv] [video]
[bibtex]
@InProceedings{Wang_2020_CVPR,
author = {Wang, Haohan and Wu, Xindi and Huang, Zeyi and Xing, Eric P.},
title = {High-Frequency Component Helps Explain the Generalization of Convolutional Neural Networks},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2020}
}