ICML 2020poster5 citations

Simple and sharp analysis of k-means||

Václav Rozhoň

Abstract

We present a simple analysis of k-means|| (Bahmani et al., PVLDB 2012) - a distributed variant of the k-means++ algorithm (Arthur and Vassilvitskii, SODA 2007). Moreover, the bound on the number of rounds is improved from $O(\log n)$ to $O(\log n / \log\log n)$, which we show to be tight.

BibTeX
@InProceedings{pmlr-v119-rozhon20a,
  title = 	 {Simple and sharp analysis of k-means||},
  author =       {Rozho{\v{n}}, V{\'a}clav},
  booktitle = 	 {Proceedings of the 37th International Conference on Machine Learning},
  pages = 	 {8266--8275},
  year = 	 {2020},
  editor = 	 {III, Hal Daumé and Singh, Aarti},
  volume = 	 {119},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {13--18 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v119/rozhon20a/rozhon20a.pdf},
  url = 	 {https://proceedings.mlr.press/v119/rozhon20a.html},
  abstract = 	 {We present a simple analysis of k-means|| (Bahmani et al., PVLDB 2012) - a distributed variant of the k-means++ algorithm (Arthur and Vassilvitskii, SODA 2007). Moreover, the bound on the number of rounds is improved from $O(\log n)$ to $O(\log n / \log\log n)$, which we show to be tight.}
}
Simple and sharp analysis of k-means|| · ICML 2020