Simple Transferability Estimation for Regression Tasks

Cuong N. Nguyen, Phong Tran, Lam Si Tung Ho, Vu Dinh, Anh T. Tran, Tal Hassner, Cuong V. Nguyen
Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence, PMLR 216:1510-1521, 2023.

Abstract

We consider transferability estimation, the problem of estimating how well deep learning models transfer from a source to a target task. We focus on regression tasks, which received little previous attention, and propose two simple and computationally efficient approaches that estimate transferability based on the negative regularized mean squared error of a linear regression model. We prove novel theoretical results connecting our approaches to the actual transferability of the optimal target models obtained from the transfer learning process. Despite their simplicity, our approaches significantly outperform existing state-of-the-art regression transferability estimators in both accuracy and efficiency. On two large-scale keypoint regression benchmarks, our approaches yield 12% to 36% better results on average while being at least 27% faster than previous state-of-the-art methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v216-nguyen23a, title = {Simple Transferability Estimation for Regression Tasks}, author = {Nguyen, Cuong N. and Tran, Phong and Ho, Lam Si Tung and Dinh, Vu and Tran, Anh T. and Hassner, Tal and Nguyen, Cuong V.}, booktitle = {Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence}, pages = {1510--1521}, year = {2023}, editor = {Evans, Robin J. and Shpitser, Ilya}, volume = {216}, series = {Proceedings of Machine Learning Research}, month = {31 Jul--04 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v216/nguyen23a/nguyen23a.pdf}, url = {https://proceedings.mlr.press/v216/nguyen23a.html}, abstract = {We consider transferability estimation, the problem of estimating how well deep learning models transfer from a source to a target task. We focus on regression tasks, which received little previous attention, and propose two simple and computationally efficient approaches that estimate transferability based on the negative regularized mean squared error of a linear regression model. We prove novel theoretical results connecting our approaches to the actual transferability of the optimal target models obtained from the transfer learning process. Despite their simplicity, our approaches significantly outperform existing state-of-the-art regression transferability estimators in both accuracy and efficiency. On two large-scale keypoint regression benchmarks, our approaches yield 12% to 36% better results on average while being at least 27% faster than previous state-of-the-art methods.} }
Endnote
%0 Conference Paper %T Simple Transferability Estimation for Regression Tasks %A Cuong N. Nguyen %A Phong Tran %A Lam Si Tung Ho %A Vu Dinh %A Anh T. Tran %A Tal Hassner %A Cuong V. Nguyen %B Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2023 %E Robin J. Evans %E Ilya Shpitser %F pmlr-v216-nguyen23a %I PMLR %P 1510--1521 %U https://proceedings.mlr.press/v216/nguyen23a.html %V 216 %X We consider transferability estimation, the problem of estimating how well deep learning models transfer from a source to a target task. We focus on regression tasks, which received little previous attention, and propose two simple and computationally efficient approaches that estimate transferability based on the negative regularized mean squared error of a linear regression model. We prove novel theoretical results connecting our approaches to the actual transferability of the optimal target models obtained from the transfer learning process. Despite their simplicity, our approaches significantly outperform existing state-of-the-art regression transferability estimators in both accuracy and efficiency. On two large-scale keypoint regression benchmarks, our approaches yield 12% to 36% better results on average while being at least 27% faster than previous state-of-the-art methods.
APA
Nguyen, C.N., Tran, P., Ho, L.S.T., Dinh, V., Tran, A.T., Hassner, T. & Nguyen, C.V.. (2023). Simple Transferability Estimation for Regression Tasks. Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 216:1510-1521 Available from https://proceedings.mlr.press/v216/nguyen23a.html.

Related Material