Worst-Case and Smoothed Analysis of the Hartigan-Wong Method for k-Means Clustering

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Manthey, Bodo, van Rhijn, Jesse
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917569151107072
author Manthey, Bodo
van Rhijn, Jesse
author_facet Manthey, Bodo
van Rhijn, Jesse
contents We analyze the running time of the Hartigan-Wong method, an old algorithm for the $k$-means clustering problem. First, we construct an instance on the line on which the method can take $2^{Ω(n)}$ steps to converge, demonstrating that the Hartigan-Wong method has exponential worst-case running time even when $k$-means is easy to solve. As this is in contrast to the empirical performance of the algorithm, we also analyze the running time in the framework of smoothed analysis. In particular, given an instance of $n$ points in $d$ dimensions, we prove that the expected number of iterations needed for the Hartigan-Wong method to terminate is bounded by $k^{12kd}\cdot poly(n, k, d, 1/σ)$ when the points in the instance are perturbed by independent $d$-dimensional Gaussian random variables of mean $0$ and standard deviation $σ$.
format Preprint
id arxiv_https___arxiv_org_abs_2309_10368
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Worst-Case and Smoothed Analysis of the Hartigan-Wong Method for k-Means Clustering
Manthey, Bodo
van Rhijn, Jesse
Data Structures and Algorithms
Computational Geometry
Probability
We analyze the running time of the Hartigan-Wong method, an old algorithm for the $k$-means clustering problem. First, we construct an instance on the line on which the method can take $2^{Ω(n)}$ steps to converge, demonstrating that the Hartigan-Wong method has exponential worst-case running time even when $k$-means is easy to solve. As this is in contrast to the empirical performance of the algorithm, we also analyze the running time in the framework of smoothed analysis. In particular, given an instance of $n$ points in $d$ dimensions, we prove that the expected number of iterations needed for the Hartigan-Wong method to terminate is bounded by $k^{12kd}\cdot poly(n, k, d, 1/σ)$ when the points in the instance are perturbed by independent $d$-dimensional Gaussian random variables of mean $0$ and standard deviation $σ$.
title Worst-Case and Smoothed Analysis of the Hartigan-Wong Method for k-Means Clustering
topic Data Structures and Algorithms
Computational Geometry
Probability
url https://arxiv.org/abs/2309.10368