Cross-Domain Feature Augmentation for Domain Generalization

TL;DR

This paper introduces XDomainMix, a novel feature augmentation technique that decomposes features into semantic components, enhancing diversity and invariance to improve domain generalization performance.

Contribution

It proposes a new feature augmentation method in the feature space that considers feature semantics, leading to improved domain generalization.

Findings

Achieves state-of-the-art results on benchmark datasets.

Facilitates learning of invariant and robust models across domains.

Enhances sample diversity through feature decomposition and augmentation.

Abstract

Domain generalization aims to develop models that are robust to distribution shifts. Existing methods focus on learning invariance across domains to enhance model robustness, and data augmentation has been widely used to learn invariant predictors, with most methods performing augmentation in the input space. However, augmentation in the input space has limited diversity whereas in the feature space is more versatile and has shown promising results. Nonetheless, feature semantics is seldom considered and existing feature augmentation methods suffer from a limited variety of augmented features. We decompose features into class-generic, class-specific, domain-generic, and domain-specific components. We propose a cross-domain feature augmentation method named XDomainMix that enables us to increase sample diversity while emphasizing the learning of invariant representations to achieve domain generalization. Experiments on widely used benchmark datasets demonstrate that our proposed method is able to achieve state-of-the-art performance. Quantitative analysis indicates that our feature augmentation approach facilitates the learning of effective models that are invariant across different domains.