Information-Maximization Clustering based on Squared-Loss Mutual Information

by   Masashi Sugiyama, et al.
Tokyo Institute of Technology

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.


Functional data clustering via information maximization

A new method for clustering functional data is proposed via information ...

Semi-Supervised Information-Maximization Clustering

Semi-supervised clustering aims to introduce prior knowledge in the deci...

Dependence Maximizing Temporal Alignment via Squared-Loss Mutual Information

The goal of temporal alignment is to establish time correspondence betwe...

Probabilistic Clustering Using Maximal Matrix Norm Couplings

In this paper, we present a local information theoretic approach to expl...

Exploiting Convexification for Bayesian Optimal Sensor Placement by Maximization of Mutual Information

Bayesian optimal sensor placement, in its full generality, seeks to maxi...

A Robust Speaker Clustering Method Based on Discrete Tied Variational Autoencoder

Recently, the speaker clustering model based on aggregation hierarchy cl...

Revisiting Discriminative Entropy Clustering and its relation to K-means

Maximization of mutual information between the model's input and output ...

Please sign up or login with your details

Forgot password? Click here to reset