Domain Generalization by Rejecting Extreme Augmentations

TL;DR

This paper introduces a simple yet effective data augmentation method for domain generalization that involves rejecting extreme transformations, leading to improved out-of-domain recognition performance.

Contribution

It proposes a novel training procedure that combines uniform sampling, stronger transformations, and rejection of extreme augmentations to enhance domain generalization.

Findings

Achieves comparable or better accuracy than state-of-the-art methods on benchmarks.

Robust improvement in out-of-domain recognition performance.

Effective rejection of harmful extreme transformations.

Abstract

Data augmentation is one of the most effective techniques for regularizing deep learning models and improving their recognition performance in a variety of tasks and domains. However, this holds for standard in-domain settings, in which the training and test data follow the same distribution. For the out-of-domain case, where the test data follow a different and unknown distribution, the best recipe for data augmentation is unclear. In this paper, we show that for out-of-domain and domain generalization settings, data augmentation can provide a conspicuous and robust improvement in performance. To do that, we propose a simple training procedure: (i) use uniform sampling on standard data augmentation transformations; (ii) increase the strength transformations to account for the higher data variance expected when working out-of-domain, and (iii) devise a new reward function to reject extreme transformations that can harm the training. With this procedure, our data augmentation scheme achieves a level of accuracy that is comparable to or better than state-of-the-art methods on benchmark domain generalization datasets. Code: https://github.com/Masseeh/DCAug