ORBi: Schnitzler François - Mixtures of Bagged Markov Tree Ensembles

Reference : Mixtures of Bagged Markov Tree Ensembles


	Document type :	Scientific congresses and symposiums : Paper published in a book
	Discipline(s) :	Engineering, computing & technology : Computer science
	To cite this reference:	http://hdl.handle.net/2268/131217

Title :	Mixtures of Bagged Markov Tree Ensembles
Language :	English
Author, co-author :	Schnitzler, François [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
	Geurts, Pierre [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
	Wehenkel, Louis [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
Publication date :	Sep-2012
Main document title :	Proceedings of the 6th European Workshop on Probabilistic Graphical Models
Editor :	Cano Utrera, Andrès
	Gómez-Olmedo, Manuel
	Nielsen, Thomas
Pages :	283-290
On invitation :	No
Audience :	International
Event name :	Sixth European Workshop on Probabilistic Graphical Models
Event date :	from 19-09-2012 to 21-09-2012
Event place (city) :	Granada
Event country :	Spain
Keywords :	[en] Bayesian networks ; Markov trees ; Perturb and Combine ; density estimation ; Chow-Liu algorithm ; Mixture of trees ; EM algorithm ; Variance reduction
Abstract :	[en] Markov trees, a probabilistic graphical model for density estimation, can be expanded in the form of a weighted average of Markov Trees. Learning these mixtures or ensembles from observations can be performed to reduce the bias or the variance of the estimated model. We propose a new combination of both, where the upper level seeks to reduce bias while the lower level seeks to reduce variance. This algorithm is evaluated empirically on datasets generated from a mixture of Markov trees and from other synthetic densities.
Research units :	systems and modelling
Funders :	Fonds pour la formation à la Recherche dans l'Industrie et dans l'Agriculture (Communauté française de Belgique) - FRIA
Target :	Researchers
Permalink :	http://hdl.handle.net/2268/131217

File(s) associated to this reference

Fulltext file(s):

	File	Commentary	Version	Size	Access
Open access	pgm12-review0.pdf		Author postprint	316.11 kB	View/Open

Additional material(s):

	File	Commentary	Version	Size	Access
Open access	presentation.pdf		-	289.55 kB	View/Open
Open access	poster1.pdf		-	313.72 kB	View/Open

All documents in ORBi are protected by a user license.

Dspace v1.5