Kernel method for clustering based on optimal target vector

TL;DR

This paper introduces the optimal target vector concept to unify supervised and unsupervised learning, constructing Ising models for clustering that depend on the entire dataset, demonstrated on iris and gene expression data.

Contribution

It presents a novel approach linking supervised and unsupervised learning through optimal target vectors and constructs data-dependent Ising models for clustering.

Findings

Effective clustering on iris dataset

Successful application to gene expression benchmarks

Models incorporate both ferro- and anti-ferromagnetic couplings

Abstract

We introduce the notion of optimal target vector, and describe how it creates a link between supervised and unsupervised learning. We exploit this notion to construct Ising models, for dichotomic clustering, whose couplings are (i) both ferro- and anti-ferromagnetic (ii) depending on the whole data-set and not only on pairs of samples. The effectiveness of the method is shown in the case of the well known iris data-set and in benchmarks of gene expression levels.