Exact Dimensionality Selection for Bayesian PCA

Abstract : We present a Bayesian model selection approach to estimate the intrinsic dimensionality of a high-dimensional dataset. To this end, we introduce a novel formulation of the probabilisitic principal component analysis model based on a normal-gamma prior distribution. In this context, we exhibit a closed-form expression of the marginal likelihood which allows to infer an optimal number of components. We also propose a heuristic based on the expected shape of the marginal likelihood curve in order to choose the hyperparameters. In non-asymptotic frameworks, we show on simulated data that this exact dimensionality selection approach is competitive with both Bayesian and frequentist state-of-the-art methods.
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01484099
Contributeur : Pierre-Alexandre Mattei <>
Soumis le : lundi 6 mars 2017 - 17:59:38
Dernière modification le : jeudi 9 mars 2017 - 01:11:21

Fichiers

ExactDimension.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01484099, version 1
  • ARXIV : 1703.02834

Collections

Citation

Charles Bouveyron, Pierre Latouche, Pierre-Alexandre Mattei. Exact Dimensionality Selection for Bayesian PCA. 2017. <hal-01484099>

Partager

Métriques

Consultations de
la notice

260

Téléchargements du document

125