The Guedon-Vershynin Semi-Definite Programming approach to low dimensional embedding for unsupervised clustering - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Frontiers in Applied Mathematics and Statistics Année : 2019

The Guedon-Vershynin Semi-Definite Programming approach to low dimensional embedding for unsupervised clustering

Résumé

This paper proposes a method for estimating the cluster matrix in the Gaussian mixture framework via Semi-Definite Programming. Theoretical error bounds are provided and a (non linear) low dimensional embedding of the data is deduced from the cluster matrix estimate. The method and its analysis is inspired by the work by Guédon and Vershynin on community detection in the stochastic block model. The adaptation is non trivial since the model is different and new Gaussian concentration arguments are needed. Our second contribution is a new Bregman-ADMM type algorithm for solving the semi-definite program and computing the embedding. This results in an efficient and scalable algorithm taking only the pairwise distances as input. The performance of the method is illustrated via Monte Carlo experiments and comparisons with other embeddings from the literature.
Fichier principal
Vignette du fichier
Clustering__GuedonVershynin_Frontiers.pdf (707.65 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02515862 , version 1 (23-03-2020)

Identifiants

Citer

Stephane Chretien, Clément Dombry, Adrien Faivre. The Guedon-Vershynin Semi-Definite Programming approach to low dimensional embedding for unsupervised clustering. Frontiers in Applied Mathematics and Statistics, 2019, ⟨10.3389/fams.2019.00041⟩. ⟨hal-02515862⟩
60 Consultations
89 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More