Semi-Supervised Hypothesis Transfer for Source-Free Domain Adaptation

TL;DR

This paper introduces a source-free semi-supervised domain adaptation method that leverages hypothesis transfer, entropy minimization, and label propagation to improve adaptation performance without source data.

Contribution

It proposes a novel hypothesis transfer approach for source-free domain adaptation using semi-supervised mutual enhancement techniques.

Findings

Achieves up to 19.9% improvement on semi-supervised adaptation tasks.

Demonstrates effectiveness across three public datasets.

Outperforms state-of-the-art methods in source-free domain adaptation.

Abstract

Domain Adaptation has been widely used to deal with the distribution shift in vision, language, multimedia etc. Most domain adaptation methods learn domain-invariant features with data from both domains available. However, such a strategy might be infeasible in practice when source data are unavailable due to data-privacy concerns. To address this issue, we propose a novel adaptation method via hypothesis transfer without accessing source data at adaptation stage. In order to fully use the limited target data, a semi-supervised mutual enhancement method is proposed, in which entropy minimization and augmented label propagation are used iteratively to perform inter-domain and intra-domain alignments. Compared with state-of-the-art methods, the experimental results on three public datasets demonstrate that our method gets up to 19.9% improvements on semi-supervised adaptation tasks.