Fast Noise Removal for k-Means Clustering

Sungjin Im, Mahshid Montazer Qaem, Benjamin Moseley, Xiaorui Sun, Rudy Zhou
; Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics, PMLR 108:456-466, 2020.

Abstract

This paper considers k-means clustering in the presence of noise. It is known that k-means clustering is highly sensitive to noise, and thus noise should be removed to obtain a quality solution. A popular formulation of this problem is called k-means clustering with outliers. The goal of k-means clustering with outliers is to discard up to a specified number z of points as noise/outliers and then find a k-means solution on the remaining data. The problem has received significant attention, yet current algorithms with theoretical guarantees suffer from either high running time or inherent loss in the solution quality. The main contribution of this paper is two-fold. Firstly, we develop a simple greedy algorithm that has provably strong worst case guarantees. The greedy algorithm adds a simple preprocessing step to remove noise, which can be combined with any k-means clustering algorithm. This algorithm gives the first pseudo-approximation-preserving reduction from k-means with outliers to k-means without outliers. Secondly, we show how to construct a coreset of size O(k log n). When combined with our greedy algorithm, we obtain a scalable, near linear time algorithm. The theoretical contributions are verified experimentally by demonstrating that the algorithm quickly removes noise and obtains a high-quality clustering.

Cite this Paper


BibTeX
@InProceedings{pmlr-v108-im20a, title = {Fast Noise Removal for k-Means Clustering}, author = {Im, Sungjin and Qaem, Mahshid Montazer and Moseley, Benjamin and Sun, Xiaorui and Zhou, Rudy}, booktitle = {Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics}, pages = {456--466}, year = {2020}, editor = {Silvia Chiappa and Roberto Calandra}, volume = {108}, series = {Proceedings of Machine Learning Research}, address = {Online}, month = {26--28 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v108/im20a/im20a.pdf}, url = {http://proceedings.mlr.press/v108/im20a.html}, abstract = {This paper considers k-means clustering in the presence of noise. It is known that k-means clustering is highly sensitive to noise, and thus noise should be removed to obtain a quality solution. A popular formulation of this problem is called k-means clustering with outliers. The goal of k-means clustering with outliers is to discard up to a specified number z of points as noise/outliers and then find a k-means solution on the remaining data. The problem has received significant attention, yet current algorithms with theoretical guarantees suffer from either high running time or inherent loss in the solution quality. The main contribution of this paper is two-fold. Firstly, we develop a simple greedy algorithm that has provably strong worst case guarantees. The greedy algorithm adds a simple preprocessing step to remove noise, which can be combined with any k-means clustering algorithm. This algorithm gives the first pseudo-approximation-preserving reduction from k-means with outliers to k-means without outliers. Secondly, we show how to construct a coreset of size O(k log n). When combined with our greedy algorithm, we obtain a scalable, near linear time algorithm. The theoretical contributions are verified experimentally by demonstrating that the algorithm quickly removes noise and obtains a high-quality clustering. } }
Endnote
%0 Conference Paper %T Fast Noise Removal for k-Means Clustering %A Sungjin Im %A Mahshid Montazer Qaem %A Benjamin Moseley %A Xiaorui Sun %A Rudy Zhou %B Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2020 %E Silvia Chiappa %E Roberto Calandra %F pmlr-v108-im20a %I PMLR %J Proceedings of Machine Learning Research %P 456--466 %U http://proceedings.mlr.press %V 108 %W PMLR %X This paper considers k-means clustering in the presence of noise. It is known that k-means clustering is highly sensitive to noise, and thus noise should be removed to obtain a quality solution. A popular formulation of this problem is called k-means clustering with outliers. The goal of k-means clustering with outliers is to discard up to a specified number z of points as noise/outliers and then find a k-means solution on the remaining data. The problem has received significant attention, yet current algorithms with theoretical guarantees suffer from either high running time or inherent loss in the solution quality. The main contribution of this paper is two-fold. Firstly, we develop a simple greedy algorithm that has provably strong worst case guarantees. The greedy algorithm adds a simple preprocessing step to remove noise, which can be combined with any k-means clustering algorithm. This algorithm gives the first pseudo-approximation-preserving reduction from k-means with outliers to k-means without outliers. Secondly, we show how to construct a coreset of size O(k log n). When combined with our greedy algorithm, we obtain a scalable, near linear time algorithm. The theoretical contributions are verified experimentally by demonstrating that the algorithm quickly removes noise and obtains a high-quality clustering.
APA
Im, S., Qaem, M.M., Moseley, B., Sun, X. & Zhou, R.. (2020). Fast Noise Removal for k-Means Clustering. Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics, in PMLR 108:456-466

Related Material