Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation

Jian Liang, Dapeng Hu, Jiashi Feng
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:6028-6039, 2020.

Abstract

Unsupervised domain adaptation (UDA) aims to leverage the knowledge learned from a labeled source dataset to solve similar tasks in a new unlabeled domain. Prior UDA methods typically require to access the source data when learning to adapt the model, making them risky and inefficient for decentralized private data. This work tackles a practical setting where only a trained source model is available and investigates how we can effectively utilize such a model without source data to solve UDA problems. We propose a simple yet generic representation learning framework, named \emph{Source HypOthesis Transfer} (SHOT). SHOT freezes the classifier module (hypothesis) of the source model and learns the target-specific feature extraction module by exploiting both information maximization and self-supervised pseudo-labeling to implicitly align representations from the target domains to the source hypothesis. To verify its versatility, we evaluate SHOT in a variety of adaptation cases including closed-set, partial-set, and open-set domain adaptation. Experiments indicate that SHOT yields state-of-the-art results among multiple domain adaptation benchmarks.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-liang20a, title = {Do We Really Need to Access the Source Data? {S}ource Hypothesis Transfer for Unsupervised Domain Adaptation}, author = {Liang, Jian and Hu, Dapeng and Feng, Jiashi}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {6028--6039}, year = {2020}, editor = {Hal Daumé III and Aarti Singh}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/liang20a/liang20a.pdf}, url = { http://proceedings.mlr.press/v119/liang20a.html }, abstract = {Unsupervised domain adaptation (UDA) aims to leverage the knowledge learned from a labeled source dataset to solve similar tasks in a new unlabeled domain. Prior UDA methods typically require to access the source data when learning to adapt the model, making them risky and inefficient for decentralized private data. This work tackles a practical setting where only a trained source model is available and investigates how we can effectively utilize such a model without source data to solve UDA problems. We propose a simple yet generic representation learning framework, named \emph{Source HypOthesis Transfer} (SHOT). SHOT freezes the classifier module (hypothesis) of the source model and learns the target-specific feature extraction module by exploiting both information maximization and self-supervised pseudo-labeling to implicitly align representations from the target domains to the source hypothesis. To verify its versatility, we evaluate SHOT in a variety of adaptation cases including closed-set, partial-set, and open-set domain adaptation. Experiments indicate that SHOT yields state-of-the-art results among multiple domain adaptation benchmarks.} }
Endnote
%0 Conference Paper %T Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation %A Jian Liang %A Dapeng Hu %A Jiashi Feng %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-liang20a %I PMLR %P 6028--6039 %U http://proceedings.mlr.press/v119/liang20a.html %V 119 %X Unsupervised domain adaptation (UDA) aims to leverage the knowledge learned from a labeled source dataset to solve similar tasks in a new unlabeled domain. Prior UDA methods typically require to access the source data when learning to adapt the model, making them risky and inefficient for decentralized private data. This work tackles a practical setting where only a trained source model is available and investigates how we can effectively utilize such a model without source data to solve UDA problems. We propose a simple yet generic representation learning framework, named \emph{Source HypOthesis Transfer} (SHOT). SHOT freezes the classifier module (hypothesis) of the source model and learns the target-specific feature extraction module by exploiting both information maximization and self-supervised pseudo-labeling to implicitly align representations from the target domains to the source hypothesis. To verify its versatility, we evaluate SHOT in a variety of adaptation cases including closed-set, partial-set, and open-set domain adaptation. Experiments indicate that SHOT yields state-of-the-art results among multiple domain adaptation benchmarks.
APA
Liang, J., Hu, D. & Feng, J.. (2020). Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:6028-6039 Available from http://proceedings.mlr.press/v119/liang20a.html .

Related Material