Nyström method with Kernel K-means++ samples as landmarks

We investigate, theoretically and empirically, the effectiveness of kernel K-means++ samples as landmarks in the Nyström method for low-rank approximation of kernel matrices. Previous empirical studies (Zhang et al., 2008; Kumar et al.,2012) observe that the landmarks obtained using (kernel) K-means...

Full description

Bibliographic Details
Main Authors: Oglic, Dino, Gaertner, Thomas
Format: Conference or Workshop Item
Published: Journal of Machine Learning Research 2017
Online Access:https://eprints.nottingham.ac.uk/43573/
_version_ 1848796718023311360
author Oglic, Dino
Gaertner, Thomas
author_facet Oglic, Dino
Gaertner, Thomas
author_sort Oglic, Dino
building Nottingham Research Data Repository
collection Online Access
description We investigate, theoretically and empirically, the effectiveness of kernel K-means++ samples as landmarks in the Nyström method for low-rank approximation of kernel matrices. Previous empirical studies (Zhang et al., 2008; Kumar et al.,2012) observe that the landmarks obtained using (kernel) K-means clustering define a good low-rank approximation of kernel matrices. However, the existing work does not provide a theoretical guarantee on the approximation error for this approach to landmark selection. We close this gap and provide the first bound on the approximation error of the Nystrom method with kernel K-means++ samples as landmarks. Moreover, for the frequently used Gaussian kernel we provide a theoretically sound motivation for performing Lloyd refinements of kernel K-means++ landmarks in the instance space. We substantiate our theoretical results empirically by comparing the approach to several state-of-the-art algorithms.
first_indexed 2025-11-14T19:52:26Z
format Conference or Workshop Item
id nottingham-43573
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:52:26Z
publishDate 2017
publisher Journal of Machine Learning Research
recordtype eprints
repository_type Digital Repository
spelling nottingham-435732020-05-04T18:59:20Z https://eprints.nottingham.ac.uk/43573/ Nyström method with Kernel K-means++ samples as landmarks Oglic, Dino Gaertner, Thomas We investigate, theoretically and empirically, the effectiveness of kernel K-means++ samples as landmarks in the Nyström method for low-rank approximation of kernel matrices. Previous empirical studies (Zhang et al., 2008; Kumar et al.,2012) observe that the landmarks obtained using (kernel) K-means clustering define a good low-rank approximation of kernel matrices. However, the existing work does not provide a theoretical guarantee on the approximation error for this approach to landmark selection. We close this gap and provide the first bound on the approximation error of the Nystrom method with kernel K-means++ samples as landmarks. Moreover, for the frequently used Gaussian kernel we provide a theoretically sound motivation for performing Lloyd refinements of kernel K-means++ landmarks in the instance space. We substantiate our theoretical results empirically by comparing the approach to several state-of-the-art algorithms. Journal of Machine Learning Research 2017-08-06 Conference or Workshop Item PeerReviewed Oglic, Dino and Gaertner, Thomas (2017) Nyström method with Kernel K-means++ samples as landmarks. In: Proceedings of the 34th International Conference on Machine Learning, 6-11 August 2017, Sydney, Australia. http://proceedings.mlr.press/v70/oglic17a
spellingShingle Oglic, Dino
Gaertner, Thomas
Nyström method with Kernel K-means++ samples as landmarks
title Nyström method with Kernel K-means++ samples as landmarks
title_full Nyström method with Kernel K-means++ samples as landmarks
title_fullStr Nyström method with Kernel K-means++ samples as landmarks
title_full_unstemmed Nyström method with Kernel K-means++ samples as landmarks
title_short Nyström method with Kernel K-means++ samples as landmarks
title_sort nyström method with kernel k-means++ samples as landmarks
url https://eprints.nottingham.ac.uk/43573/
https://eprints.nottingham.ac.uk/43573/