Implicit Differentiation for Hyperparameter Tuning the Weighted Graphical Lasso - Inria - Institut national de recherche en sciences et technologies du numérique
Conference Papers Year : 2023

Implicit Differentiation for Hyperparameter Tuning the Weighted Graphical Lasso

Abstract

We provide a framework and algorithm for tuning the hyperparameters of the Graphical Lasso via a bilevel optimization problem solved with a first-order method. In particular, we derive the Jacobian of the Graphical Lasso solution with respect to its regularization hyperparameters.
Fichier principal
Vignette du fichier
main.pdf (550.25 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04151796 , version 1 (05-07-2023)

Licence

Identifiers

  • HAL Id : hal-04151796 , version 1

Cite

Can Pouliquen, Paulo Gonçalves, Mathurin Massias, Titouan Vayer. Implicit Differentiation for Hyperparameter Tuning the Weighted Graphical Lasso. GRETSI 2023 - XXIXème Colloque Francophone de Traitement du Signal et des Images, Aug 2023, Grenoble (France), France. pp.1-4. ⟨hal-04151796⟩
147 View
110 Download

Share

More