Statistical analysis and parameter selection for Mapper

Mathieu Carriere 1 Bertrand Michel 2 Steve Y. Oudot 1
1 DATASHAPE - Understanding the Shape of Data
CRISAM - Inria Sophia Antipolis - Méditerranée , Inria Saclay - Ile de France
Abstract : In this article, we study the question of the statistical convergence of the 1-dimensional Mapper to its continuous analogue, the Reeb graph. We show that the Mapper is an optimal estimator of the Reeb graph, which gives, as a byproduct, a method to automatically tune its parameters and compute confidence regions on its topological features, such as its loops and flares. This allows to circumvent the issue of testing a large grid of parameters and keeping the most stable ones in the brute-force setting, which is widely used in visualization, clustering and feature selection with the Mapper.
Type de document :
Article dans une revue
Journal of Machine Learning Research, Journal of Machine Learning Research, 2018
Liste complète des métadonnées

Littérature citée [11 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01633106
Contributeur : Steve Oudot <>
Soumis le : mercredi 10 octobre 2018 - 17:37:18
Dernière modification le : samedi 13 octobre 2018 - 01:19:30

Fichier

Mapper-Stat.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01633106, version 2

Citation

Mathieu Carriere, Bertrand Michel, Steve Y. Oudot. Statistical analysis and parameter selection for Mapper. Journal of Machine Learning Research, Journal of Machine Learning Research, 2018. 〈hal-01633106v2〉

Partager

Métriques

Consultations de la notice

25

Téléchargements de fichiers

55