Adapting $k$-means algorithms for outliers

TL;DR

This paper extends classical $k$-means algorithms to effectively handle outliers, providing stronger theoretical guarantees and improved outlier output control in both sequential and distributed settings.

Contribution

It introduces new adaptations of $k$-means algorithms for outliers with stronger approximation guarantees and precise outlier output control, improving upon previous methods.

Findings

Algorithms output $(1+ ext{epsilon})z$ outliers

Achieve $O(1/ ext{epsilon})$-approximation to the objective

Matching lower bound of $ ext{Omega}(nk^2/z)$ in the oracle model

Abstract

This paper shows how to adapt several simple and classical sampling-based algorithms for the $k$-means problem to the setting with outliers. Recently, Bhaskara et al. (NeurIPS 2019) showed how to adapt the classical $k$-means++ algorithm to the setting with outliers. However, their algorithm needs to output $O(\log (k) \cdot z)$ outliers, where $z$ is the number of true outliers, to match the $O(\log k)$-approximation guarantee of $k$-means++. In this paper, we build on their ideas and show how to adapt several sequential and distributed $k$-means algorithms to the setting with outliers, but with substantially stronger theoretical guarantees: our algorithms output $(1+\varepsilon)z$ outliers while achieving an $O(1 / \varepsilon)$-approximation to the objective function. In the sequential world, we achieve this by adapting a recent algorithm of Lattanzi and Sohler (ICML 2019). In the distributed setting, we adapt a simple algorithm of Guha et al. (IEEE Trans. Know. and Data Engineering 2003) and the popular $k$-means$\|$ of Bahmani et al. (PVLDB 2012). A theoretical application of our techniques is an algorithm with running time $\tilde{O}(nk^2/z)$ that achieves an $O(1)$-approximation to the objective function while outputting $O(z)$ outliers, assuming $k \ll z \ll n$. This is complemented with a matching lower bound of $\Omega(nk^2/z)$ for this problem in the oracle model.