Implicit Differentiation for Hyperparameter Tuning the Weighted Graphical Lasso
Machine Learning
2024-06-14 v1 Optimization and Control
Machine Learning
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.
Cite
@article{arxiv.2307.02130,
title = {Implicit Differentiation for Hyperparameter Tuning the Weighted Graphical Lasso},
author = {Can Pouliquen and Paulo Gonçalves and Mathurin Massias and Titouan Vayer},
journal= {arXiv preprint arXiv:2307.02130},
year = {2024}
}