Fast Detection of Block Boundaries in Block-Wise Constant Matrices - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Fast Detection of Block Boundaries in Block-Wise Constant Matrices

Résumé

We propose a novel approach for estimating the location of block boundaries (change-points) in a random matrix consisting of a block wise constant matrix observed in white noise. Our method consists in rephrasing this task as a variable selection issue. We use a penalized least-squares criterion with an 1-type penalty for dealing with this problem. We first provide some theoretical results ensuring the consistency of our change-point estimators. Then, we explain how to implement our method in a very efficient way. Finally, we provide some empirical evidence to support our claims and apply our approach to data coming from molecular biology which can be used for better understanding the structure of the chromatin.
Fichier principal
Vignette du fichier
Brault2016bFast.pdf (4.94 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01574362 , version 1 (14-08-2017)

Licence

Paternité

Identifiants

Citer

Vincent Brault, Julien Chiquet, Céline Lévy-Leduc. Fast Detection of Block Boundaries in Block-Wise Constant Matrices. 12th International Conference on Machine Learning and Data Mining in Pattern Recognition (MLDM 2016), Petra Perner, Jul 2016, New York, NY, United States. pp.214-228, ⟨10.1007/978-3-319-41920-6_16⟩. ⟨hal-01574362⟩
138 Consultations
119 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More