SDCA-Powered Inexact Dual Augmented Lagrangian Method for Fast CRF Learning - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

SDCA-Powered Inexact Dual Augmented Lagrangian Method for Fast CRF Learning

(1) , (1)
1

Résumé

We propose an efficient dual augmented La-grangian formulation to learn conditional random fields (CRF). Our algorithm, which can be interpreted as an inexact gradient descent algorithm on the multiplier, does not require to perform global inference iteratively, and requires only a fixed number of stochastic clique-wise updates at each epoch to obtain a sufficiently good estimate of the gradient w.r.t. the Lagrange multipliers. We prove that the proposed algorithm enjoys global linear convergence for both the primal and the dual objectives. Our experiments show that the proposed algorithm outperforms state-of-the-art baselines in terms of speed of convergence.
Fichier principal
Vignette du fichier
hu18-combined.pdf (1.06 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-01754043 , version 1 (30-03-2018)

Identifiants

  • HAL Id : hal-01754043 , version 1

Citer

Shell Xu Hu, Guillaume Obozinski. SDCA-Powered Inexact Dual Augmented Lagrangian Method for Fast CRF Learning. 21st International Conference on Artificial Intelligence and Statistics (AISTATS), Apr 2018, Lanzarote, Spain. ⟨hal-01754043⟩
87 Consultations
31 Téléchargements

Partager

Gmail Facebook Twitter LinkedIn More