Towards Accurate Model Selection in Deep Unsupervised Domain Adaptation

Kaichao You, Ximei Wang, Mingsheng Long, Michael Jordan
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:7124-7133, 2019.

Abstract

Deep unsupervised domain adaptation (Deep UDA) methods successfully leverage rich labeled data in a source domain to boost the performance on related but unlabeled data in a target domain. However, algorithm comparison is cumbersome in Deep UDA due to the absence of accurate and standardized model selection method, posing an obstacle to further advances in the field. Existing model selection methods for Deep UDA are either highly biased, restricted, unstable, or even controversial (requiring labeled target data). To this end, we propose Deep Embedded Validation (DEV), which embeds adapted feature representation into the validation procedure to obtain unbiased estimation of the target risk with bounded variance. The variance is further reduced by the technique of control variate. The efficacy of the method has been justified both theoretically and empirically.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-you19a, title = {Towards Accurate Model Selection in Deep Unsupervised Domain Adaptation}, author = {You, Kaichao and Wang, Ximei and Long, Mingsheng and Jordan, Michael}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {7124--7133}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/you19a/you19a.pdf}, url = {https://proceedings.mlr.press/v97/you19a.html}, abstract = {Deep unsupervised domain adaptation (Deep UDA) methods successfully leverage rich labeled data in a source domain to boost the performance on related but unlabeled data in a target domain. However, algorithm comparison is cumbersome in Deep UDA due to the absence of accurate and standardized model selection method, posing an obstacle to further advances in the field. Existing model selection methods for Deep UDA are either highly biased, restricted, unstable, or even controversial (requiring labeled target data). To this end, we propose Deep Embedded Validation (DEV), which embeds adapted feature representation into the validation procedure to obtain unbiased estimation of the target risk with bounded variance. The variance is further reduced by the technique of control variate. The efficacy of the method has been justified both theoretically and empirically.} }
Endnote
%0 Conference Paper %T Towards Accurate Model Selection in Deep Unsupervised Domain Adaptation %A Kaichao You %A Ximei Wang %A Mingsheng Long %A Michael Jordan %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-you19a %I PMLR %P 7124--7133 %U https://proceedings.mlr.press/v97/you19a.html %V 97 %X Deep unsupervised domain adaptation (Deep UDA) methods successfully leverage rich labeled data in a source domain to boost the performance on related but unlabeled data in a target domain. However, algorithm comparison is cumbersome in Deep UDA due to the absence of accurate and standardized model selection method, posing an obstacle to further advances in the field. Existing model selection methods for Deep UDA are either highly biased, restricted, unstable, or even controversial (requiring labeled target data). To this end, we propose Deep Embedded Validation (DEV), which embeds adapted feature representation into the validation procedure to obtain unbiased estimation of the target risk with bounded variance. The variance is further reduced by the technique of control variate. The efficacy of the method has been justified both theoretically and empirically.
APA
You, K., Wang, X., Long, M. & Jordan, M.. (2019). Towards Accurate Model Selection in Deep Unsupervised Domain Adaptation. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:7124-7133 Available from https://proceedings.mlr.press/v97/you19a.html.

Related Material