Domain Generalization via Adversarially Learned Novel Domains

Abstract

This study focuses on the domain generalization task, which aims to learn a model that generalizes to unseen domains by utilizing multiple training domains. More specifically, we follow the idea of adversarial data augmentation, which aims to synthesize and augment training data with “hard” domains to improve the model’s domain generalization ability. However, previous studies augmented training data only with samples similar to the training data, resulting in limited generalization ability. To alleviate this issue, we propose a novel adversarial data augmentation method, termed GADA (generative adversarial domain augmentation), which employs an image-to-image translation model to obtain a distribution of novel domains that are semantically different from the training domains, and, at the same time, hard to classify. Evaluation and further analysis suggest that GADA fits our expectation; adversarial data augmentation with semantically different samples leads to better domain generalization performance.

Keywords

• Adversarial training
• data augmentation
• domain generalization