Incremental Randomized Sketching for Online Kernel Learning

Xiao Zhang, Shizhong Liao
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:7394-7403, 2019.

Abstract

Randomized sketching has been used in offline kernel learning, but it cannot be applied directly to online kernel learning due to the lack of incremental maintenances for randomized sketches with regret guarantees. To address these issues, we propose a novel incremental randomized sketching approach for online kernel learning, which has efficient incremental maintenances with theoretical guarantees. We construct two incremental randomized sketches using the sparse transform matrix and the sampling matrix for kernel matrix approximation, update the incremental randomized sketches using rank-$1$ modifications, and construct an time-varying explicit feature mapping for online kernel learning. We prove that the proposed incremental randomized sketching is statistically unbiased for the matrix product approximation, obtains a $1 + \epsilon$ relative-error bound for the kernel matrix approximation, enjoys a sublinear regret bound for online kernel learning, and has constant time and space complexities at each round for incremental maintenances. Experimental results demonstrate that the incremental randomized sketching achieves a better learning performance in terms of accuracy and efficiency even in adversarial environments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-zhang19h, title = {Incremental Randomized Sketching for Online Kernel Learning}, author = {Zhang, Xiao and Liao, Shizhong}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {7394--7403}, 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/zhang19h/zhang19h.pdf}, url = {https://proceedings.mlr.press/v97/zhang19h.html}, abstract = {Randomized sketching has been used in offline kernel learning, but it cannot be applied directly to online kernel learning due to the lack of incremental maintenances for randomized sketches with regret guarantees. To address these issues, we propose a novel incremental randomized sketching approach for online kernel learning, which has efficient incremental maintenances with theoretical guarantees. We construct two incremental randomized sketches using the sparse transform matrix and the sampling matrix for kernel matrix approximation, update the incremental randomized sketches using rank-$1$ modifications, and construct an time-varying explicit feature mapping for online kernel learning. We prove that the proposed incremental randomized sketching is statistically unbiased for the matrix product approximation, obtains a $1 + \epsilon$ relative-error bound for the kernel matrix approximation, enjoys a sublinear regret bound for online kernel learning, and has constant time and space complexities at each round for incremental maintenances. Experimental results demonstrate that the incremental randomized sketching achieves a better learning performance in terms of accuracy and efficiency even in adversarial environments.} }
Endnote
%0 Conference Paper %T Incremental Randomized Sketching for Online Kernel Learning %A Xiao Zhang %A Shizhong Liao %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-zhang19h %I PMLR %P 7394--7403 %U https://proceedings.mlr.press/v97/zhang19h.html %V 97 %X Randomized sketching has been used in offline kernel learning, but it cannot be applied directly to online kernel learning due to the lack of incremental maintenances for randomized sketches with regret guarantees. To address these issues, we propose a novel incremental randomized sketching approach for online kernel learning, which has efficient incremental maintenances with theoretical guarantees. We construct two incremental randomized sketches using the sparse transform matrix and the sampling matrix for kernel matrix approximation, update the incremental randomized sketches using rank-$1$ modifications, and construct an time-varying explicit feature mapping for online kernel learning. We prove that the proposed incremental randomized sketching is statistically unbiased for the matrix product approximation, obtains a $1 + \epsilon$ relative-error bound for the kernel matrix approximation, enjoys a sublinear regret bound for online kernel learning, and has constant time and space complexities at each round for incremental maintenances. Experimental results demonstrate that the incremental randomized sketching achieves a better learning performance in terms of accuracy and efficiency even in adversarial environments.
APA
Zhang, X. & Liao, S.. (2019). Incremental Randomized Sketching for Online Kernel Learning. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:7394-7403 Available from https://proceedings.mlr.press/v97/zhang19h.html.

Related Material