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.
|Number of pages||12|
|Journal||Proceedings of Machine Learning Research|
|State||Published - 2023|
|Event||39th Conference on Uncertainty in Artificial Intelligence, UAI 2023 - Pittsburgh, United States|
Duration: 31 Jul 2023 → 4 Aug 2023
Bibliographical noteFunding Information:
LSTH was supported by the Canada Research Chairs program, the NSERC Discovery Grant RGPIN-2018-05447, and the NSERC Discovery Launch Supplement DGECR-2018-00181. VD was supported by the University of Delaware Research Foundation (UDRF) Strategic Initiatives Grant, and the National Science Foundation Grant DMS-1951474.
© UAI 2023. All rights reserved.