Adversarial Invariant Feature Learning with Accuracy Constraint for Domain Generalization

TL;DR

This paper introduces AFLAC, a novel adversarial feature learning method that balances domain invariance and classification accuracy, improving domain generalization performance on multiple datasets.

Contribution

It proposes the concept of accuracy-constrained domain invariance and develops AFLAC to explicitly optimize this balance during training.

Findings

AFLAC outperforms existing domain-invariance methods on synthetic and real datasets.

The method effectively manages the trade-off between invariance and accuracy.

Empirical results validate the importance of considering class-domain dependency.

Abstract

Learning domain-invariant representation is a dominant approach for domain generalization (DG), where we need to build a classifier that is robust toward domain shifts. However, previous domain-invariance-based methods overlooked the underlying dependency of classes on domains, which is responsible for the trade-off between classification accuracy and domain invariance. Because the primary purpose of DG is to classify unseen domains rather than the invariance itself, the improvement of the invariance can negatively affect DG performance under this trade-off. To overcome the problem, this study first expands the analysis of the trade-off by Xie et. al., and provides the notion of accuracy-constrained domain invariance, which means the maximum domain invariance within a range that does not interfere with accuracy. We then propose a novel method adversarial feature learning with accuracy constraint (AFLAC), which explicitly leads to that invariance on adversarial training. Empirical validations show that the performance of AFLAC is superior to that of domain-invariance-based methods on both synthetic and three real-world datasets, supporting the importance of considering the dependency and the efficacy of the proposed method.