Robust Covariate Shift Regression

Xiangli Chen, Mathew Monfort, Anqi Liu, Brian D. Ziebart
Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, PMLR 51:1270-1279, 2016.

Abstract

In many learning settings, the source data available to train a regression model differs from the target data it encounters when making predictions due to input distribution shift. Appropriately dealing with this situation remains an important challenge. Existing methods attempt to “reweight” the source data samples to better represent the target domain, but this introduces strong inductive biases that are highly extrapolative and can often err greatly in practice. We propose a robust approach for regression under covariate shift that embraces the uncertainty resulting from sample selection bias by producing regression models that are explicitly robust to it. We demonstrate the benefits of our approach on a number of regression tasks.

Cite this Paper


BibTeX
@InProceedings{pmlr-v51-chen16d, title = {Robust Covariate Shift Regression}, author = {Chen, Xiangli and Monfort, Mathew and Liu, Anqi and Ziebart, Brian D.}, booktitle = {Proceedings of the 19th International Conference on Artificial Intelligence and Statistics}, pages = {1270--1279}, year = {2016}, editor = {Gretton, Arthur and Robert, Christian C.}, volume = {51}, series = {Proceedings of Machine Learning Research}, address = {Cadiz, Spain}, month = {09--11 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v51/chen16d.pdf}, url = {https://proceedings.mlr.press/v51/chen16d.html}, abstract = {In many learning settings, the source data available to train a regression model differs from the target data it encounters when making predictions due to input distribution shift. Appropriately dealing with this situation remains an important challenge. Existing methods attempt to “reweight” the source data samples to better represent the target domain, but this introduces strong inductive biases that are highly extrapolative and can often err greatly in practice. We propose a robust approach for regression under covariate shift that embraces the uncertainty resulting from sample selection bias by producing regression models that are explicitly robust to it. We demonstrate the benefits of our approach on a number of regression tasks.} }
Endnote
%0 Conference Paper %T Robust Covariate Shift Regression %A Xiangli Chen %A Mathew Monfort %A Anqi Liu %A Brian D. Ziebart %B Proceedings of the 19th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2016 %E Arthur Gretton %E Christian C. Robert %F pmlr-v51-chen16d %I PMLR %P 1270--1279 %U https://proceedings.mlr.press/v51/chen16d.html %V 51 %X In many learning settings, the source data available to train a regression model differs from the target data it encounters when making predictions due to input distribution shift. Appropriately dealing with this situation remains an important challenge. Existing methods attempt to “reweight” the source data samples to better represent the target domain, but this introduces strong inductive biases that are highly extrapolative and can often err greatly in practice. We propose a robust approach for regression under covariate shift that embraces the uncertainty resulting from sample selection bias by producing regression models that are explicitly robust to it. We demonstrate the benefits of our approach on a number of regression tasks.
RIS
TY - CPAPER TI - Robust Covariate Shift Regression AU - Xiangli Chen AU - Mathew Monfort AU - Anqi Liu AU - Brian D. Ziebart BT - Proceedings of the 19th International Conference on Artificial Intelligence and Statistics DA - 2016/05/02 ED - Arthur Gretton ED - Christian C. Robert ID - pmlr-v51-chen16d PB - PMLR DP - Proceedings of Machine Learning Research VL - 51 SP - 1270 EP - 1279 L1 - http://proceedings.mlr.press/v51/chen16d.pdf UR - https://proceedings.mlr.press/v51/chen16d.html AB - In many learning settings, the source data available to train a regression model differs from the target data it encounters when making predictions due to input distribution shift. Appropriately dealing with this situation remains an important challenge. Existing methods attempt to “reweight” the source data samples to better represent the target domain, but this introduces strong inductive biases that are highly extrapolative and can often err greatly in practice. We propose a robust approach for regression under covariate shift that embraces the uncertainty resulting from sample selection bias by producing regression models that are explicitly robust to it. We demonstrate the benefits of our approach on a number of regression tasks. ER -
APA
Chen, X., Monfort, M., Liu, A. & Ziebart, B.D.. (2016). Robust Covariate Shift Regression. Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 51:1270-1279 Available from https://proceedings.mlr.press/v51/chen16d.html.

Related Material