Source-Free Domain Adaptation via Multi-view Contrastive Learning

TL;DR

This paper introduces a novel source-free domain adaptation method that uses multi-view contrastive learning and sample memory to improve pseudo-label quality and prototype representation, achieving significant accuracy gains.

Contribution

It proposes a three-phase approach combining reliable sample memory, multi-view contrastive learning, and noisy label filtering for enhanced source-free domain adaptation.

Findings

Achieves about 2% higher accuracy than the second-best method.

Attains an average of 6% improvement over 13 state-of-the-art approaches.

Demonstrates effectiveness on three benchmark datasets.

Abstract

Domain adaptation has become a widely adopted approach in machine learning due to the high costs associated with labeling data. It is typically applied when access to a labeled source domain is available. However, in real-world scenarios, privacy concerns often restrict access to sensitive information, such as fingerprints, bank account details, and facial images. A promising solution to this issue is Source-Free Unsupervised Domain Adaptation (SFUDA), which enables domain adaptation without requiring access to labeled target domain data. Recent research demonstrates that SFUDA can effectively address domain discrepancies; however, two key challenges remain: (1) the low quality of prototype samples, and (2) the incorrect assignment of pseudo-labels. To tackle these challenges, we propose a method consisting of three main phases. In the first phase, we introduce a Reliable Sample Memory (RSM) module to improve the quality of prototypes by selecting more representative samples. In the second phase, we employ a Multi-View Contrastive Learning (MVCL) approach to enhance pseudo-label quality by leveraging multiple data augmentations. In the final phase, we apply a noisy label filtering technique to further refine the pseudo-labels. Our experiments on three benchmark datasets - VisDA 2017, Office-Home, and Office-31 - demonstrate that our method achieves approximately 2 percent and 6 percent improvements in classification accuracy over the second-best method and the average of 13 well-known state-of-the-art approaches, respectively.