Inferring large graphs using l1-penalized likelihood - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Statistics and Computing Année : 2017

Inferring large graphs using l1-penalized likelihood

Résumé

We address the issue of recovering the structure of large sparse directed acyclic graphs from noisy observations of the system. We propose a novel procedure based on a specific formulation of the l1-norm regularized maximum likelihood, which decomposes the graph estimation into two optimization sub-problems: topological structure and node order learning. We provide oracle inequalities for the graph estimator, as well as an algorithm to solve the induced optimization problem, in the form of a convex program embedded in a genetic algorithm. We apply our method to various data sets (including data from the DREAM4 challenge) and show that it compares favorably to state-of-the-art methods.
Fichier principal
Vignette du fichier
HALMChampion2.pdf (579.27 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01172745 , version 1 (07-07-2015)
hal-01172745 , version 2 (24-01-2017)
hal-01172745 , version 3 (04-10-2017)

Identifiants

Citer

Magali Champion, Victor Picheny, Matthieu Vignes. Inferring large graphs using l1-penalized likelihood. Statistics and Computing, 2017, ⟨10.1198/jasa.2011.ap10346⟩. ⟨hal-01172745v3⟩
193 Consultations
352 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More