Showing 1–1 of 1 results for author: Mazza-Anthony, C
-
Learning Gaussian Graphical Models with Ordered Weighted L1 Regularization
Authors:
Cody Mazza-Anthony,
Bogdan Mazoure,
Mark Coates
Abstract:
We address the task of identifying densely connected subsets of multivariate Gaussian random variables within a graphical model framework. We propose two novel estimators based on the Ordered Weighted $\ell_1$ (OWL) norm: 1) The Graphical OWL (GOWL) is a penalized likelihood method that applies the OWL norm to the lower triangle components of the precision matrix. 2) The column-by-column Graphical…
▽ More
We address the task of identifying densely connected subsets of multivariate Gaussian random variables within a graphical model framework. We propose two novel estimators based on the Ordered Weighted $\ell_1$ (OWL) norm: 1) The Graphical OWL (GOWL) is a penalized likelihood method that applies the OWL norm to the lower triangle components of the precision matrix. 2) The column-by-column Graphical OWL (ccGOWL) estimates the precision matrix by performing OWL regularized linear regressions. Both methods can simultaneously identify highly correlated groups of variables and control the sparsity in the resulting precision matrix. We formulate GOWL such that it solves a composite optimization problem and establish that the estimator has a unique global solution. In addition, we prove sufficient grou** conditions for each column of the ccGOWL precision matrix estimate. We propose proximal descent algorithms to find the optimum for both estimators. For synthetic data where group structure is present, the ccGOWL estimator requires significantly reduced computation and achieves similar or greater accuracy than state-of-the-art estimators. Timing comparisons are presented and demonstrates the superior computational efficiency of the ccGOWL. We illustrate the grou** performance of the ccGOWL method on a cancer gene expression data set and an equities data set.
△ Less
Submitted 19 November, 2020; v1 submitted 6 June, 2019;
originally announced June 2019.
