Provable Guarantees for Gradient-Based Meta-Learning

Maria-Florina Balcan, Mikhail Khodak, Ameet Talwalkar
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:424-433, 2019.

Abstract

We study the problem of meta-learning through the lens of online convex optimization, developing a meta-algorithm bridging the gap between popular gradient-based meta-learning and classical regularization-based multi-task transfer methods. Our method is the first to simultaneously satisfy good sample efficiency guarantees in the convex setting, with generalization bounds that improve with task-similarity, while also being computationally scalable to modern deep learning architectures and the many-task setting. Despite its simplicity, the algorithm matches, up to a constant factor, a lower bound on the performance of any such parameter-transfer method under natural task similarity assumptions. We use experiments in both convex and deep learning settings to verify and demonstrate the applicability of our theory.

Cite this Paper


BibTeX
@InProceedings{pmlr-v97-balcan19a, title = {Provable Guarantees for Gradient-Based Meta-Learning}, author = {Balcan, Maria-Florina and Khodak, Mikhail and Talwalkar, Ameet}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {424--433}, year = {2019}, editor = {Kamalika Chaudhuri and Ruslan Salakhutdinov}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/balcan19a/balcan19a.pdf}, url = { http://proceedings.mlr.press/v97/balcan19a.html }, abstract = {We study the problem of meta-learning through the lens of online convex optimization, developing a meta-algorithm bridging the gap between popular gradient-based meta-learning and classical regularization-based multi-task transfer methods. Our method is the first to simultaneously satisfy good sample efficiency guarantees in the convex setting, with generalization bounds that improve with task-similarity, while also being computationally scalable to modern deep learning architectures and the many-task setting. Despite its simplicity, the algorithm matches, up to a constant factor, a lower bound on the performance of any such parameter-transfer method under natural task similarity assumptions. We use experiments in both convex and deep learning settings to verify and demonstrate the applicability of our theory.} }
Endnote
%0 Conference Paper %T Provable Guarantees for Gradient-Based Meta-Learning %A Maria-Florina Balcan %A Mikhail Khodak %A Ameet Talwalkar %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-balcan19a %I PMLR %P 424--433 %U http://proceedings.mlr.press/v97/balcan19a.html %V 97 %X We study the problem of meta-learning through the lens of online convex optimization, developing a meta-algorithm bridging the gap between popular gradient-based meta-learning and classical regularization-based multi-task transfer methods. Our method is the first to simultaneously satisfy good sample efficiency guarantees in the convex setting, with generalization bounds that improve with task-similarity, while also being computationally scalable to modern deep learning architectures and the many-task setting. Despite its simplicity, the algorithm matches, up to a constant factor, a lower bound on the performance of any such parameter-transfer method under natural task similarity assumptions. We use experiments in both convex and deep learning settings to verify and demonstrate the applicability of our theory.
APA
Balcan, M., Khodak, M. & Talwalkar, A.. (2019). Provable Guarantees for Gradient-Based Meta-Learning. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:424-433 Available from http://proceedings.mlr.press/v97/balcan19a.html .

Related Material