Semi-Supervised Hypothesis Transfer for Source-Free Domain Adaptation
This addresses domain adaptation for vision, language, and multimedia applications where data privacy restricts access to source data, representing an incremental advance over existing methods.
The paper tackles the problem of domain adaptation when source data is unavailable due to privacy concerns by proposing a semi-supervised hypothesis transfer method, achieving up to 19.9% improvement on semi-supervised tasks compared to state-of-the-art methods.
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.