A Better k-means++ Algorithm via Local Search

Silvio Lattanzi, Christian Sohler
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:3662-3671, 2019.

Abstract

In this paper, we develop a new variant of k-means++ seeding that in expectation achieves a constant approximation guarantee. We obtain this result by a simple combination of k-means++ sampling with a local search strategy. We evaluate our algorithm empirically and show that it also improves the quality of a solution in practice.

Cite this Paper


BibTeX
@InProceedings{pmlr-v97-lattanzi19a, title = {A Better k-means++ Algorithm via Local Search}, author = {Lattanzi, Silvio and Sohler, Christian}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {3662--3671}, 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/lattanzi19a/lattanzi19a.pdf}, url = {http://proceedings.mlr.press/v97/lattanzi19a.html}, abstract = {In this paper, we develop a new variant of k-means++ seeding that in expectation achieves a constant approximation guarantee. We obtain this result by a simple combination of k-means++ sampling with a local search strategy. We evaluate our algorithm empirically and show that it also improves the quality of a solution in practice.} }
Endnote
%0 Conference Paper %T A Better k-means++ Algorithm via Local Search %A Silvio Lattanzi %A Christian Sohler %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-lattanzi19a %I PMLR %P 3662--3671 %U http://proceedings.mlr.press/v97/lattanzi19a.html %V 97 %X In this paper, we develop a new variant of k-means++ seeding that in expectation achieves a constant approximation guarantee. We obtain this result by a simple combination of k-means++ sampling with a local search strategy. We evaluate our algorithm empirically and show that it also improves the quality of a solution in practice.
APA
Lattanzi, S. & Sohler, C.. (2019). A Better k-means++ Algorithm via Local Search. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:3662-3671 Available from http://proceedings.mlr.press/v97/lattanzi19a.html.

Related Material