Adapting to Latent Subgroup Shifts via Concepts and Proxies

Ibrahim Alabdulmohsin, Nicole Chiou, Alexander D’Amour, Arthur Gretton, Sanmi Koyejo, Matt J. Kusner, Stephen R. Pfohl, Olawale Salaudeen, Jessica Schrouff, Katherine Tsai
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:9637-9661, 2023.

Abstract

We address the problem of unsupervised domain adaptation when the source domain differs from the target domain because of a shift in the distribution of a latent subgroup. When this subgroup confounds all observed data, neither covariate shift nor label shift assumptions apply. We show that the optimal target predictor can be non-parametrically identified with the help of concept and proxy variables available only in the source domain, and unlabeled data from the target. The identification results are constructive, immediately suggesting an algorithm for estimating the optimal predictor in the target. For continuous observations, when this algorithm becomes impractical, we propose a latent variable model specific to the data generation process at hand. We show how the approach degrades as the size of the shift changes, and verify that it outperforms both covariate and label shift adjustment.

Cite this Paper

BibTeX
@InProceedings{pmlr-v206-alabdulmohsin23a, title = {Adapting to Latent Subgroup Shifts via Concepts and Proxies}, author = {Alabdulmohsin, Ibrahim and Chiou, Nicole and D'Amour, Alexander and Gretton, Arthur and Koyejo, Sanmi and Kusner, Matt J. and Pfohl, Stephen R. and Salaudeen, Olawale and Schrouff, Jessica and Tsai, Katherine}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {9637--9661}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/alabdulmohsin23a/alabdulmohsin23a.pdf}, url = {https://proceedings.mlr.press/v206/alabdulmohsin23a.html}, abstract = {We address the problem of unsupervised domain adaptation when the source domain differs from the target domain because of a shift in the distribution of a latent subgroup. When this subgroup confounds all observed data, neither covariate shift nor label shift assumptions apply. We show that the optimal target predictor can be non-parametrically identified with the help of concept and proxy variables available only in the source domain, and unlabeled data from the target. The identification results are constructive, immediately suggesting an algorithm for estimating the optimal predictor in the target. For continuous observations, when this algorithm becomes impractical, we propose a latent variable model specific to the data generation process at hand. We show how the approach degrades as the size of the shift changes, and verify that it outperforms both covariate and label shift adjustment.} }
Endnote
%0 Conference Paper %T Adapting to Latent Subgroup Shifts via Concepts and Proxies %A Ibrahim Alabdulmohsin %A Nicole Chiou %A Alexander D’Amour %A Arthur Gretton %A Sanmi Koyejo %A Matt J. Kusner %A Stephen R. Pfohl %A Olawale Salaudeen %A Jessica Schrouff %A Katherine Tsai %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-alabdulmohsin23a %I PMLR %P 9637--9661 %U https://proceedings.mlr.press/v206/alabdulmohsin23a.html %V 206 %X We address the problem of unsupervised domain adaptation when the source domain differs from the target domain because of a shift in the distribution of a latent subgroup. When this subgroup confounds all observed data, neither covariate shift nor label shift assumptions apply. We show that the optimal target predictor can be non-parametrically identified with the help of concept and proxy variables available only in the source domain, and unlabeled data from the target. The identification results are constructive, immediately suggesting an algorithm for estimating the optimal predictor in the target. For continuous observations, when this algorithm becomes impractical, we propose a latent variable model specific to the data generation process at hand. We show how the approach degrades as the size of the shift changes, and verify that it outperforms both covariate and label shift adjustment.
APA
Alabdulmohsin, I., Chiou, N., D’Amour, A., Gretton, A., Koyejo, S., Kusner, M.J., Pfohl, S.R., Salaudeen, O., Schrouff, J. & Tsai, K.. (2023). Adapting to Latent Subgroup Shifts via Concepts and Proxies. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 206:9637-9661 Available from https://proceedings.mlr.press/v206/alabdulmohsin23a.html.

Related Material