Continuous Relaxation of MAP Inference: A Nonconvex Perspective

Abstract : In this paper, we study a nonconvex continuous relaxation of MAP inference in discrete Markov random fields (MRFs). We show that for arbitrary MRFs, this relaxation is tight, and a discrete stationary point of it can be easily reached by a simple block coordinate descent algorithm. In addition, we study the resolution of this relaxation using popular gradient methods, and further propose a more effective solution using a multilinear decomposition framework based on the alternating direction method of multi-pliers (ADMM). Experiments on many real-world problems demonstrate that the proposed ADMM significantly outper-forms other nonconvex relaxation based methods, and compares favorably with state of the art MRF optimization algorithms in different settings.
Type de document :
Communication dans un congrès
CVPR 2018 - IEEE Conference on Computer Vision and Pattern Recognition, Jun 2018, Salt Lake City, United States. pp.1-19, 〈http://cvpr2018.thecvf.com/〉
Liste complète des métadonnées

Littérature citée [13 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01716514
Contributeur : D. Khuê Lê-Huu <>
Soumis le : lundi 26 février 2018 - 16:44:57
Dernière modification le : vendredi 6 avril 2018 - 13:32:01
Document(s) archivé(s) le : lundi 28 mai 2018 - 20:33:29

Fichier

norelax_cvpr2018.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01716514, version 2

Citation

D. Khuê Lê-Huu, Nikos Paragios. Continuous Relaxation of MAP Inference: A Nonconvex Perspective. CVPR 2018 - IEEE Conference on Computer Vision and Pattern Recognition, Jun 2018, Salt Lake City, United States. pp.1-19, 〈http://cvpr2018.thecvf.com/〉. 〈hal-01716514v2〉

Partager

Métriques

Consultations de la notice

202

Téléchargements de fichiers

46