Exploiting Domain-Specific Features to Enhance Domain Generalization

TL;DR

This paper introduces a novel framework called mDSDI that disentangles domain-invariant and domain-specific features using meta-learning, improving generalization to unseen domains in domain generalization tasks.

Contribution

It extends beyond invariance by jointly learning domain-specific and invariant features, demonstrating improved generalization through a theoretically grounded approach.

Findings

mDSDI achieves competitive results with state-of-the-art methods.

Domain-specific features are crucial for better generalization.

Ablation study confirms the importance of domain-specific information.

Abstract

Domain Generalization (DG) aims to train a model, from multiple observed source domains, in order to perform well on unseen target domains. To obtain the generalization capability, prior DG approaches have focused on extracting domain-invariant information across sources to generalize on target domains, while useful domain-specific information which strongly correlates with labels in individual domains and the generalization to target domains is usually ignored. In this paper, we propose meta-Domain Specific-Domain Invariant (mDSDI) - a novel theoretically sound framework that extends beyond the invariance view to further capture the usefulness of domain-specific information. Our key insight is to disentangle features in the latent space while jointly learning both domain-invariant and domain-specific features in a unified framework. The domain-specific representation is optimized through the meta-learning framework to adapt from source domains, targeting a robust generalization on unseen domains. We empirically show that mDSDI provides competitive results with state-of-the-art techniques in DG. A further ablation study with our generated dataset, Background-Colored-MNIST, confirms the hypothesis that domain-specific is essential, leading to better results when compared with only using domain-invariant.