A Bregman Method for Structure Learning on Sparse Directed Acyclic Graphs - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2021

A Bregman Method for Structure Learning on Sparse Directed Acyclic Graphs

Résumé

We develop a Bregman proximal gradient method for structure learning on linear structural causal models. While the problem is non-convex, has high curvature and is in fact NP-hard, Bregman gradient methods allow us to neutralize at least part of the impact of curvature by measuring smoothness against a highly nonlinear kernel. This allows the method to make longer steps and significantly improves convergence. Each iteration requires solving a Bregman proximal step which is convex and efficiently solvable for our particular choice of kernel. We test our method on various synthetic and real data sets.

Dates et versions

hal-03165618 , version 1 (10-03-2021)

Identifiants

Citer

Manon Romain, Alexandre d'Aspremont. A Bregman Method for Structure Learning on Sparse Directed Acyclic Graphs. 2021. ⟨hal-03165618⟩
2825 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More