Guardado en:
Detalles Bibliográficos
Autores principales: Knoll, Christian, Weller, Adrian, Pernkopf, Franz
Formato: Preprint
Publicado: 2018
Materias:
Acceso en línea:https://arxiv.org/abs/1812.01339
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914995257737216
author Knoll, Christian
Weller, Adrian
Pernkopf, Franz
author_facet Knoll, Christian
Weller, Adrian
Pernkopf, Franz
contents Belief propagation (BP) is a popular method for performing probabilistic inference on graphical models. In this work, we enhance BP and propose self-guided belief propagation (SBP) that incorporates the pairwise potentials only gradually. This homotopy continuation method converges to a unique solution and increases the accuracy without increasing the computational burden. We provide a formal analysis to demonstrate that SBP finds the global optimum of the Bethe approximation for attractive models where all variables favor the same state. Moreover, we apply SBP to various graphs with random potentials and empirically show that: (i) SBP is superior in terms of accuracy whenever BP converges, and (ii) SBP obtains a unique, stable, and accurate solution whenever BP does not converge.
format Preprint
id arxiv_https___arxiv_org_abs_1812_01339
institution arXiv
publishDate 2018
record_format arxiv
spellingShingle Self-Guided Belief Propagation -- A Homotopy Continuation Method
Knoll, Christian
Weller, Adrian
Pernkopf, Franz
Machine Learning
Belief propagation (BP) is a popular method for performing probabilistic inference on graphical models. In this work, we enhance BP and propose self-guided belief propagation (SBP) that incorporates the pairwise potentials only gradually. This homotopy continuation method converges to a unique solution and increases the accuracy without increasing the computational burden. We provide a formal analysis to demonstrate that SBP finds the global optimum of the Bethe approximation for attractive models where all variables favor the same state. Moreover, we apply SBP to various graphs with random potentials and empirically show that: (i) SBP is superior in terms of accuracy whenever BP converges, and (ii) SBP obtains a unique, stable, and accurate solution whenever BP does not converge.
title Self-Guided Belief Propagation -- A Homotopy Continuation Method
topic Machine Learning
url https://arxiv.org/abs/1812.01339