Unified Transfer Learning in High-Dimensional Linear Regression

Shuo Shuo Liu
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, PMLR 238:1036-1044, 2024.

Abstract

Transfer learning plays a key role in modern data analysis when: (1) the target data are scarce but the source data are sufficient; (2) the distributions of the source and target data are heterogeneous. This paper develops an interpretable unified transfer learning model, termed as UTrans, which can detect both transferable variables and source data. More specifically, we establish the estimation error bounds and prove that our bounds are lower than those with target data only. Besides, we propose a source detection algorithm based on hypothesis testing to exclude the nontransferable data. We evaluate and compare UTrans to the existing algorithms in multiple experiments. It is shown that UTrans attains much lower estimation and prediction errors than the existing methods, while preserving interpretability. We finally apply it to the US intergenerational mobility data and compare our proposed algorithms to the classical machine learning algorithms.

Cite this Paper


BibTeX
@InProceedings{pmlr-v238-shuo-liu24a, title = {Unified Transfer Learning in High-Dimensional Linear Regression}, author = {Shuo Liu, Shuo}, booktitle = {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics}, pages = {1036--1044}, year = {2024}, editor = {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen}, volume = {238}, series = {Proceedings of Machine Learning Research}, month = {02--04 May}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v238/shuo-liu24a/shuo-liu24a.pdf}, url = {https://proceedings.mlr.press/v238/shuo-liu24a.html}, abstract = {Transfer learning plays a key role in modern data analysis when: (1) the target data are scarce but the source data are sufficient; (2) the distributions of the source and target data are heterogeneous. This paper develops an interpretable unified transfer learning model, termed as UTrans, which can detect both transferable variables and source data. More specifically, we establish the estimation error bounds and prove that our bounds are lower than those with target data only. Besides, we propose a source detection algorithm based on hypothesis testing to exclude the nontransferable data. We evaluate and compare UTrans to the existing algorithms in multiple experiments. It is shown that UTrans attains much lower estimation and prediction errors than the existing methods, while preserving interpretability. We finally apply it to the US intergenerational mobility data and compare our proposed algorithms to the classical machine learning algorithms.} }
Endnote
%0 Conference Paper %T Unified Transfer Learning in High-Dimensional Linear Regression %A Shuo Shuo Liu %B Proceedings of The 27th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2024 %E Sanjoy Dasgupta %E Stephan Mandt %E Yingzhen Li %F pmlr-v238-shuo-liu24a %I PMLR %P 1036--1044 %U https://proceedings.mlr.press/v238/shuo-liu24a.html %V 238 %X Transfer learning plays a key role in modern data analysis when: (1) the target data are scarce but the source data are sufficient; (2) the distributions of the source and target data are heterogeneous. This paper develops an interpretable unified transfer learning model, termed as UTrans, which can detect both transferable variables and source data. More specifically, we establish the estimation error bounds and prove that our bounds are lower than those with target data only. Besides, we propose a source detection algorithm based on hypothesis testing to exclude the nontransferable data. We evaluate and compare UTrans to the existing algorithms in multiple experiments. It is shown that UTrans attains much lower estimation and prediction errors than the existing methods, while preserving interpretability. We finally apply it to the US intergenerational mobility data and compare our proposed algorithms to the classical machine learning algorithms.
APA
Shuo Liu, S.. (2024). Unified Transfer Learning in High-Dimensional Linear Regression. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 238:1036-1044 Available from https://proceedings.mlr.press/v238/shuo-liu24a.html.

Related Material