Fast Bayesian Network Structure Learning using Quasi-Determinism Screening

Thibaud Rahier 1, 2 Sylvain Marié 1 Stéphane Girard 2 Florence Forbes 2
2 MISTIS - Modelling and Inference of Complex and Structured Stochastic Systems
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : Learning the structure of Bayesian networks from data is a NP-Hard problem that involves optimization over a super-exponential sized space. In this work, we show that in most real life datasets, a number of the arcs contained in the final structure can be pre-screened at low computational cost with a limited impact on the global graph score. We formalize the identification of these arcs via the notion of quasi-determinism, and propose an associated algorithm that narrows the structure learning task down to a subset of the original variables. We show, on diverse benchmark datasets, that this algorithm exhibits a significant decrease in computational time and complexity for only a little decrease in performance score.
Type de document :
Pré-publication, Document de travail
2018
Liste complète des métadonnées

Littérature citée [31 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01691217
Contributeur : Thibaud Rahier <>
Soumis le : lundi 12 mars 2018 - 14:14:51
Dernière modification le : mercredi 11 avril 2018 - 01:59:15
Document(s) archivé(s) le : mercredi 13 juin 2018 - 13:55:29

Fichier

FastBNSL_V2 2.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01691217, version 3

Collections

Citation

Thibaud Rahier, Sylvain Marié, Stéphane Girard, Florence Forbes. Fast Bayesian Network Structure Learning using Quasi-Determinism Screening. 2018. 〈hal-01691217v3〉

Partager

Métriques

Consultations de la notice

339

Téléchargements de fichiers

133