Discriminative Domain-Invariant Adversarial Network for Deep Domain Generalization

TL;DR

This paper introduces DDIAN, a novel adversarial network that enhances domain generalization by learning discriminative and domain-invariant features, outperforming existing methods on benchmark datasets.

Contribution

The paper proposes a discriminative domain-invariant adversarial network (DDIAN) that addresses limitations of previous methods by ensuring feature discriminativeness and robust domain invariance.

Findings

DDIAN achieves superior accuracy on unseen target domains.

Extensive experiments validate the effectiveness of DDIAN over state-of-the-art approaches.

The method improves generalization by combining discriminative features with domain alignment.

Abstract

Domain generalization approaches aim to learn a domain invariant prediction model for unknown target domains from multiple training source domains with different distributions. Significant efforts have recently been committed to broad domain generalization, which is a challenging and topical problem in machine learning and computer vision communities. Most previous domain generalization approaches assume that the conditional distribution across the domains remain the same across the source domains and learn a domain invariant model by minimizing the marginal distributions. However, the assumption of a stable conditional distribution of the training source domains does not really hold in practice. The hyperplane learned from the source domains will easily misclassify samples scattered at the boundary of clusters or far from their corresponding class centres. To address the above two drawbacks, we propose a discriminative domain-invariant adversarial network (DDIAN) for domain generalization. The discriminativeness of the features are guaranteed through a discriminative feature module and domain-invariant features are guaranteed through the global domain and local sub-domain alignment modules. Extensive experiments on several benchmarks show that DDIAN achieves better prediction on unseen target data during training compared to state-of-the-art domain generalization approaches.