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.
@article{arxiv.2209.02369,
title = {Improving Robustness to Out-of-Distribution Data by Frequency-based Augmentation},
author = {Koki Mukai and Soichiro Kumano and Toshihiko Yamasaki},
journal= {arXiv preprint arXiv:2209.02369},
year = {2022}
}