Information-Maximization Clustering based on Squared-Loss Mutual Information

TL;DR

This paper introduces a new clustering method based on squared-loss mutual information that simplifies the optimization process, providing an analytical solution via kernel eigenvalue decomposition and enabling effective model selection.

Contribution

It proposes a novel squared-loss mutual information-based clustering method that offers an analytical solution and practical model selection, improving over existing non-convex approaches.

Findings

Analytical clustering solution via kernel eigenvalue decomposition

Effective model selection procedure for tuning parameters

Demonstrated usefulness through experiments

Abstract

Information-maximization clustering learns a probabilistic classifier in an unsupervised manner so that mutual information between feature vectors and cluster assignments is maximized. A notable advantage of this approach is that it only involves continuous optimization of model parameters, which is substantially easier to solve than discrete optimization of cluster assignments. However, existing methods still involve non-convex optimization problems, and therefore finding a good local optimal solution is not straightforward in practice. In this paper, we propose an alternative information-maximization clustering method based on a squared-loss variant of mutual information. This novel approach gives a clustering solution analytically in a computationally efficient way via kernel eigenvalue decomposition. Furthermore, we provide a practical model selection procedure that allows us to objectively optimize tuning parameters included in the kernel function. Through experiments, we demonstrate the usefulness of the proposed approach.