Domain Consistency Regularization for Unsupervised Multi-source Domain Adaptive Classification

TL;DR

This paper introduces CRMA, a novel end-to-end network for multi-source unsupervised domain adaptation that aligns multiple domains and improves classification accuracy by regularizing domain consistency.

Contribution

The paper proposes a new domain consistency regularization method that aligns multiple source domains with the target and among themselves, addressing a gap in existing MUDA approaches.

Findings

CRMA outperforms existing MUDA methods on multiple datasets.

It effectively aligns domain distributions and decision boundaries.

The adaptive authority strategy improves pseudo label accuracy.

Abstract

Deep learning-based multi-source unsupervised domain adaptation (MUDA) has been actively studied in recent years. Compared with single-source unsupervised domain adaptation (SUDA), domain shift in MUDA exists not only between the source and target domains but also among multiple source domains. Most existing MUDA algorithms focus on extracting domain-invariant representations among all domains whereas the task-specific decision boundaries among classes are largely neglected. In this paper, we propose an end-to-end trainable network that exploits domain Consistency Regularization for unsupervised Multi-source domain Adaptive classification (CRMA). CRMA aligns not only the distributions of each pair of source and target domains but also that of all domains. For each pair of source and target domains, we employ an intra-domain consistency to regularize a pair of domain-specific classifiers to achieve intra-domain alignment. In addition, we design an inter-domain consistency that targets joint inter-domain alignment among all domains. To address different similarities between multiple source domains and the target domain, we design an authorization strategy that assigns different authorities to domain-specific classifiers adaptively for optimal pseudo label prediction and self-training. Extensive experiments show that CRMA tackles unsupervised domain adaptation effectively under a multi-source setup and achieves superior adaptation consistently across multiple MUDA datasets.