Min-Sum Clustering (with Outliers)

TL;DR

This paper presents a polynomial-time pseudo-approximation algorithm for min-sum clustering with outliers, allowing exclusion of a small fraction of points and providing bicriteria guarantees in general metric and Euclidean spaces.

Contribution

It introduces a new bicriteria approximation algorithm for min-sum clustering with outliers that works in arbitrary metric and Euclidean spaces, with guarantees depending on the outlier fraction.

Findings

Clusters at least (1-ε) n' points with cost within a constant factor of optimal

Algorithm works in polynomial time for arbitrary metric and Euclidean spaces

Provides bicriteria approximation with guarantees depending on 1/ε

Abstract

We give a constant factor polynomial time pseudo-approximation algorithm for min-sum clustering with or without outliers. The algorithm is allowed to exclude an arbitrarily small constant fraction of the points. For instance, we show how to compute a solution that clusters 98\% of the input data points and pays no more than a constant factor times the optimal solution that clusters 99\% of the input data points. More generally, we give the following bicriteria approximation: For any $\eps > 0$, for any instance with $n$ input points and for any positive integer $n'\le n$, we compute in polynomial time a clustering of at least $(1-\eps) n'$ points of cost at most a constant factor greater than the optimal cost of clustering $n'$ points. The approximation guarantee grows with $\frac{1}{\eps}$. Our results apply to instances of points in real space endowed with squared Euclidean distance, as well as to points in a metric space, where the number of clusters, and also the dimension if relevant, is arbitrary (part of the input, not an absolute constant).