Multi-Source Unsupervised Domain Adaptation with Prototype Aggregation

TL;DR

This paper introduces a prototype aggregation approach for multi-source unsupervised domain adaptation, addressing class discrepancy, noisy pseudo-labels, and transferability to improve model performance across domains.

Contribution

It proposes a novel method modeling class and domain discrepancies with prototypes, incorporating transferability quantification and alignment strategies for better adaptation.

Findings

Outperforms most state-of-the-art methods on three benchmarks.

Effectively quantifies class-specific discrepancies using pseudo-labels.

Provides interpretable analysis of the adaptation process.

Abstract

Multi-source domain adaptation (MSDA) plays an important role in industrial model generalization. Recent efforts on MSDA focus on enhancing multi-domain distributional alignment while omitting three issues, e.g., the class-level discrepancy quantification, the unavailability of noisy pseudo-label, and source transferability discrimination, potentially resulting in suboptimal adaption performance. Therefore, we address these issues by proposing a prototype aggregation method that models the discrepancy between source and target domains at the class and domain levels. Our method achieves domain adaptation based on a group of prototypes (i.e., representative feature embeddings). A similarity score-based strategy is designed to quantify the transferability of each domain. At the class level, our method quantifies class-specific cross-domain discrepancy according to reliable target pseudo-labels. At the domain level, our method establishes distributional alignment between noisy pseudo-labeled target samples and the source domain prototypes. Therefore, adaptation at the class and domain levels establishes a complementary mechanism to obtain accurate predictions. The results on three standard benchmarks demonstrate that our method outperforms most state-of-the-art methods. In addition, we provide further elaboration of the proposed method in light of the interpretable results obtained from the analysis experiments.