A Fast Approximation Scheme for Low-Dimensional $k$-Means
Abstract
We consider the popular -means problem in -dimensional Euclidean space. Recently Friggstad, Rezapour, Salavatipour [FOCS'16] and Cohen-Addad, Klein, Mathieu [FOCS'16] showed that the standard local search algorithm yields a -approximation in time , giving the first polynomial-time approximation scheme for the problem in low-dimensional Euclidean space. While local search achieves optimal approximation guarantees, it is not competitive with the state-of-the-art heuristics such as the famous -means++ and -sampling algorithms. In this paper, we aim at bridging the gap between theory and practice by giving a -approximation algorithm for low-dimensional -means running in time , and so matching the running time of the -means++ and -sampling heuristics up to polylogarithmic factors. We speed-up the local search approach by making a non-standard use of randomized dissections that allows to find the best local move efficiently using a quite simple dynamic program. We hope that our techniques could help design better local search heuristics for geometric problems. We note that the doubly exponential dependency on is necessary as -means is APX-hard in dimension .
Cite
@article{arxiv.1708.07381,
title = {A Fast Approximation Scheme for Low-Dimensional $k$-Means},
author = {Vincent Cohen-Addad},
journal= {arXiv preprint arXiv:1708.07381},
year = {2017}
}