A Constant-Factor Bi-Criteria Approximation Guarantee for K-Means++

Abstract

This paper studies the $k$-means++ algorithm for clustering as well as the class of $D^\ell$ sampling algorithms to which $k$-means++ belongs. It is shown that for any constant factor $\beta > 1$, selecting $\beta k$ cluster centers by $D^\ell$ sampling yields a constant-factor approximation to the optimal clustering with $k$ centers, in expectation and without conditions on the dataset. This result extends the previously known $O(\log k)$ guarantee for the case $\beta = 1$ to the constant-factor bi-criteria regime. It also improves upon an existing constant-factor bi-criteria result that holds only with constant probability.

Cite

Text

Wei. "A Constant-Factor Bi-Criteria Approximation Guarantee for K-Means++." Neural Information Processing Systems, 2016.

Markdown

[Wei. "A Constant-Factor Bi-Criteria Approximation Guarantee for K-Means++." Neural Information Processing Systems, 2016.](https://mlanthology.org/neurips/2016/wei2016neurips-constantfactor/)

BibTeX

@inproceedings{wei2016neurips-constantfactor,
  title     = {{A Constant-Factor Bi-Criteria Approximation Guarantee for K-Means++}},
  author    = {Wei, Dennis},
  booktitle = {Neural Information Processing Systems},
  year      = {2016},
  pages     = {604-612},
  url       = {https://mlanthology.org/neurips/2016/wei2016neurips-constantfactor/}
}