Flexible Priors for Exemplar-based Clustering

TL;DR

This paper introduces a novel way to incorporate flexible priors, including Dirichlet process priors, into exemplar-based clustering methods like affinity propagation, improving their ability to recover true data clusters.

Contribution

It develops a new model integrating priors into affinity propagation and presents an efficient belief propagation algorithm for enhanced clustering performance.

Findings

Better recovery of true clusters with prior integration

Efficient belief propagation algorithm developed

Expanded range of clustering priors improves results

Abstract

Exemplar-based clustering methods have been shown to produce state-of-the-art results on a number of synthetic and real-world clustering problems. They are appealing because they offer computational benefits over latent-mean models and can handle arbitrary pairwise similarity measures between data points. However, when trying to recover underlying structure in clustering problems, tailored similarity measures are often not enough; we also desire control over the distribution of cluster sizes. Priors such as Dirichlet process priors allow the number of clusters to be unspecified while expressing priors over data partitions. To our knowledge, they have not been applied to exemplar-based models. We show how to incorporate priors, including Dirichlet process priors, into the recently introduced affinity propagation algorithm. We develop an efficient maxproduct belief propagation algorithm for our new model and demonstrate experimentally how the expanded range of clustering priors allows us to better recover true clusterings in situations where we have some information about the generating process.