English

Fast and Accurate $k$-means++ via Rejection Sampling

Machine Learning 2020-12-23 v1 Data Structures and Algorithms

Abstract

kk-means++ \cite{arthur2007k} is a widely used clustering algorithm that is easy to implement, has nice theoretical guarantees and strong empirical performance. Despite its wide adoption, kk-means++ sometimes suffers from being slow on large data-sets so a natural question has been to obtain more efficient algorithms with similar guarantees. In this paper, we present a near linear time algorithm for kk-means++ seeding. Interestingly our algorithm obtains the same theoretical guarantees as kk-means++ and significantly improves earlier results on fast kk-means++ seeding. Moreover, we show empirically that our algorithm is significantly faster than kk-means++ and obtains solutions of equivalent quality.

Keywords

Cite

@article{arxiv.2012.11891,
  title  = {Fast and Accurate $k$-means++ via Rejection Sampling},
  author = {Vincent Cohen-Addad and Silvio Lattanzi and Ashkan Norouzi-Fard and Christian Sohler and Ola Svensson},
  journal= {arXiv preprint arXiv:2012.11891},
  year   = {2020}
}
R2 v1 2026-06-23T21:11:35.591Z