English

A Refined Approximation for Euclidean k-Means

Data Structures and Algorithms 2021-09-21 v2 Computational Geometry

Abstract

In the Euclidean kk-Means problem we are given a collection of nn points DD in an Euclidean space and a positive integer kk. Our goal is to identify a collection of kk points in the same space (centers) so as to minimize the sum of the squared Euclidean distances between each point in DD and the closest center. This problem is known to be APX-hard and the current best approximation ratio is a primal-dual 6.3576.357 approximation based on a standard LP for the problem [Ahmadian et al. FOCS'17, SICOMP'20]. In this note we show how a minor modification of Ahmadian et al.'s analysis leads to a slightly improved 6.129036.12903 approximation. As a related result, we also show that the mentioned LP has integrality gap at least 16+515>1.2157\frac{16+\sqrt{5}}{15}>1.2157.

Keywords

Cite

@article{arxiv.2107.07358,
  title  = {A Refined Approximation for Euclidean k-Means},
  author = {Fabrizio Grandoni and Rafail Ostrovsky and Yuval Rabani and Leonard J. Schulman and Rakesh Venkat},
  journal= {arXiv preprint arXiv:2107.07358},
  year   = {2021}
}

Comments

Corrected a confusing typo in a formula on page 5 and added one remark

R2 v1 2026-06-24T04:13:53.347Z