Kernel k-Groups via Hartigan's Method

TL;DR

This paper introduces kernel k-groups, a clustering method based on energy statistics in negative type metric spaces, extending Hartigan's method to kernel spaces with improved performance and applications in community detection.

Contribution

It develops a novel kernel clustering algorithm called kernel k-groups, extending Hartigan's method to kernel spaces and demonstrating its efficiency and effectiveness in high-dimensional and community detection tasks.

Findings

Kernel k-groups outperform spectral clustering in experiments.

The method is computationally comparable to kernel k-means.

Effective in community detection in sparse stochastic block models.

Abstract

Energy statistics was proposed by Sz\' ekely in the 80's inspired by Newton's gravitational potential in classical mechanics and it provides a model-free hypothesis test for equality of distributions. In its original form, energy statistics was formulated in Euclidean spaces. More recently, it was generalized to metric spaces of negative type. In this paper, we consider a formulation for the clustering problem using a weighted version of energy statistics in spaces of negative type. We show that this approach leads to a quadratically constrained quadratic program in the associated kernel space, establishing connections with graph partitioning problems and kernel methods in machine learning. To find local solutions of such an optimization problem, we propose kernel k-groups, which is an extension of Hartigan's method to kernel spaces. Kernel k-groups is cheaper than spectral clustering and has the same computational cost as kernel k-means (which is based on Lloyd's heuristic) but our numerical results show an improved performance, especially in higher dimensions. Moreover, we verify the efficiency of kernel k-groups in community detection in sparse stochastic block models which has fascinating applications in several areas of science.