Improving Robustness to out-of-Distribution Data by Frequency-Based Augmentation
Lists
Only you see this
Abstract
Although Convolutional Neural Networks (CNNs) have high accuracy in image recognition, they are vulnerable to adversarial examples and out-of-distribution data, and the difference from human recognition has been pointed out. In order to improve the robustness against out-of-distribution data, we present a frequency-based data augmentation technique that replaces the frequency components with other images of the same class. When the training data are CIFAR10 and the out-of-distribution data are SVHN, the Area Under Receiver Operating Characteristic (AUROC) curve of the model trained with the proposed method increases from 89.22% to 98.15%, and further increased to 98.59% when combined with another data augmentation method. Furthermore, we experimentally demonstrate that the robust model for out-of-distribution data uses a lot of high-frequency components of the image.
BibTeX
@inproceedings{mukai2022improving,
author = {Koki Mukai and Soichiro Kumano and Toshihiko Yamasaki},
title = {{Improving Robustness to out-of-Distribution Data by Frequency-Based Augmentation}},
booktitle = {ICIP},
year = {2022},
doi = {10.1109/icip46576.2022.9897504},
}Loading related content...