Online Multi-Kernel Learning with Graph-Structured Feedback

Pouya M Ghari, Yanning Shen
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:3474-3483, 2020.

Abstract

Multi-kernel learning (MKL) exhibits reliable performance in nonlinear function approximation tasks. Instead of using one kernel, it learns the optimal kernel from a pre-selected dictionary of kernels. The selection of the dictionary has crucial impact on both the performance and complexity of MKL. Specifically, inclusion of a large number of irrelevant kernels may impair the accuracy, and increase the complexity of MKL algorithms. To enhance the accuracy, and alleviate the computational burden, the present paper develops a novel scheme which actively chooses relevant kernels. The proposed framework models the pruned kernel combination as feedback collected from a graph, that is refined ’on the fly.’ Leveraging the random feature approximation, we propose an online scalable multi-kernel learning approach with graph feedback, and prove that the proposed algorithm enjoys sublinear regret. Numerical tests on real datasets demonstrate the effectiveness of the novel approach.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-ghari20a, title = {Online Multi-Kernel Learning with Graph-Structured Feedback}, author = {Ghari, Pouya M and Shen, Yanning}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {3474--3483}, year = {2020}, editor = {Hal Daumé III and Aarti Singh}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/ghari20a/ghari20a.pdf}, url = { http://proceedings.mlr.press/v119/ghari20a.html }, abstract = {Multi-kernel learning (MKL) exhibits reliable performance in nonlinear function approximation tasks. Instead of using one kernel, it learns the optimal kernel from a pre-selected dictionary of kernels. The selection of the dictionary has crucial impact on both the performance and complexity of MKL. Specifically, inclusion of a large number of irrelevant kernels may impair the accuracy, and increase the complexity of MKL algorithms. To enhance the accuracy, and alleviate the computational burden, the present paper develops a novel scheme which actively chooses relevant kernels. The proposed framework models the pruned kernel combination as feedback collected from a graph, that is refined ’on the fly.’ Leveraging the random feature approximation, we propose an online scalable multi-kernel learning approach with graph feedback, and prove that the proposed algorithm enjoys sublinear regret. Numerical tests on real datasets demonstrate the effectiveness of the novel approach.} }
Endnote
%0 Conference Paper %T Online Multi-Kernel Learning with Graph-Structured Feedback %A Pouya M Ghari %A Yanning Shen %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-ghari20a %I PMLR %P 3474--3483 %U http://proceedings.mlr.press/v119/ghari20a.html %V 119 %X Multi-kernel learning (MKL) exhibits reliable performance in nonlinear function approximation tasks. Instead of using one kernel, it learns the optimal kernel from a pre-selected dictionary of kernels. The selection of the dictionary has crucial impact on both the performance and complexity of MKL. Specifically, inclusion of a large number of irrelevant kernels may impair the accuracy, and increase the complexity of MKL algorithms. To enhance the accuracy, and alleviate the computational burden, the present paper develops a novel scheme which actively chooses relevant kernels. The proposed framework models the pruned kernel combination as feedback collected from a graph, that is refined ’on the fly.’ Leveraging the random feature approximation, we propose an online scalable multi-kernel learning approach with graph feedback, and prove that the proposed algorithm enjoys sublinear regret. Numerical tests on real datasets demonstrate the effectiveness of the novel approach.
APA
Ghari, P.M. & Shen, Y.. (2020). Online Multi-Kernel Learning with Graph-Structured Feedback. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:3474-3483 Available from http://proceedings.mlr.press/v119/ghari20a.html .

Related Material