Learning What and Where to Transfer

Yunhun Jang, Hankook Lee, Sung Ju Hwang, Jinwoo Shin
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:3030-3039, 2019.

Abstract

As the application of deep learning has expanded to real-world problems with insufficient volume of training data, transfer learning recently has gained much attention as means of improving the performance in such small-data regime. However, when existing methods are applied between heterogeneous architectures and tasks, it becomes more important to manage their detailed configurations and often requires exhaustive tuning on them for the desired performance. To address the issue, we propose a novel transfer learning approach based on meta-learning that can automatically learn what knowledge to transfer from the source network to where in the target network. Given source and target networks, we propose an efficient training scheme to learn meta-networks that decide (a) which pairs of layers between the source and target networks should be matched for knowledge transfer and (b) which features and how much knowledge from each feature should be transferred. We validate our meta-transfer approach against recent transfer learning methods on various datasets and network architectures, on which our automated scheme significantly outperforms the prior baselines that find “what and where to transfer” in a hand-crafted manner.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-jang19b, title = {Learning What and Where to Transfer}, author = {Jang, Yunhun and Lee, Hankook and Hwang, Sung Ju and Shin, Jinwoo}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {3030--3039}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/jang19b/jang19b.pdf}, url = {https://proceedings.mlr.press/v97/jang19b.html}, abstract = {As the application of deep learning has expanded to real-world problems with insufficient volume of training data, transfer learning recently has gained much attention as means of improving the performance in such small-data regime. However, when existing methods are applied between heterogeneous architectures and tasks, it becomes more important to manage their detailed configurations and often requires exhaustive tuning on them for the desired performance. To address the issue, we propose a novel transfer learning approach based on meta-learning that can automatically learn what knowledge to transfer from the source network to where in the target network. Given source and target networks, we propose an efficient training scheme to learn meta-networks that decide (a) which pairs of layers between the source and target networks should be matched for knowledge transfer and (b) which features and how much knowledge from each feature should be transferred. We validate our meta-transfer approach against recent transfer learning methods on various datasets and network architectures, on which our automated scheme significantly outperforms the prior baselines that find “what and where to transfer” in a hand-crafted manner.} }
Endnote
%0 Conference Paper %T Learning What and Where to Transfer %A Yunhun Jang %A Hankook Lee %A Sung Ju Hwang %A Jinwoo Shin %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-jang19b %I PMLR %P 3030--3039 %U https://proceedings.mlr.press/v97/jang19b.html %V 97 %X As the application of deep learning has expanded to real-world problems with insufficient volume of training data, transfer learning recently has gained much attention as means of improving the performance in such small-data regime. However, when existing methods are applied between heterogeneous architectures and tasks, it becomes more important to manage their detailed configurations and often requires exhaustive tuning on them for the desired performance. To address the issue, we propose a novel transfer learning approach based on meta-learning that can automatically learn what knowledge to transfer from the source network to where in the target network. Given source and target networks, we propose an efficient training scheme to learn meta-networks that decide (a) which pairs of layers between the source and target networks should be matched for knowledge transfer and (b) which features and how much knowledge from each feature should be transferred. We validate our meta-transfer approach against recent transfer learning methods on various datasets and network architectures, on which our automated scheme significantly outperforms the prior baselines that find “what and where to transfer” in a hand-crafted manner.
APA
Jang, Y., Lee, H., Hwang, S.J. & Shin, J.. (2019). Learning What and Where to Transfer. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:3030-3039 Available from https://proceedings.mlr.press/v97/jang19b.html.

Related Material