Modèles graphiques paramétriques pour la modélisation des lois de génération dans des processus de branchement multitypes

Pierre Fernique 1, 2 Jean-Baptiste Durand 1, 3 Yann Guédon 1, 2
1 VIRTUAL PLANTS - Modeling plant morphogenesis at different scales, from genes to phenotype
CRISAM - Inria Sophia Antipolis - Méditerranée , INRA - Institut National de la Recherche Agronomique, UMR AGAP - Amélioration génétique et adaptation des plantes méditerranéennes et tropicales
3 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 : We address discrete-state models for tree-structured data. Our aim is to introduce parametric multitype branching processes that can be effciently estimated on the basis of data of limited size. Each generation distribution is modeled by a mixture of graphical models. Their estimation relies on selection of a conditional independence graph and selection of the number of components of the mixture model. We show on apple tree flowering data that this framework allows us to identify tree patterns corresponding to a more or less pronounced alternation of flowering, depending on cultivar. Keywords: Graphical mixture model, multitype branching process, multivariate discrete distribution ; tree pattern.
Complete list of metadatas

Cited literature [6 references]  Display  Hide  Download

https://hal.inria.fr/hal-01058313
Contributor : Jean-Baptiste Durand <>
Submitted on : Tuesday, August 26, 2014 - 3:28:51 PM
Last modification on : Tuesday, April 16, 2019 - 1:32:06 AM
Long-term archiving on : Thursday, November 27, 2014 - 4:16:39 PM

File

jds2014_fernique.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01058313, version 1

Collections

Citation

Pierre Fernique, Jean-Baptiste Durand, Yann Guédon. Modèles graphiques paramétriques pour la modélisation des lois de génération dans des processus de branchement multitypes. 46èmes Journées de Statistique, Jun 2014, Rennes, France. ⟨hal-01058313⟩

Share

Metrics

Record views

698

Files downloads

180