Improving Robustness to Out-of-Distribution Data by Frequency-based Augmentation
This addresses the vulnerability of CNNs to out-of-distribution data, which is a critical issue for reliable image recognition systems, but the approach is incremental as it builds on existing augmentation methods.
The paper tackles the problem of improving robustness to out-of-distribution data in CNNs by proposing a frequency-based data augmentation technique that replaces frequency components with other images of the same class, resulting in an increase in AUROC from 89.22% to 98.15% on CIFAR10 vs. SVHN, and up to 98.59% when combined with another method.
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.