A Pathwise Algorithm for Covariance Selection

2009/08/02 by Krishnamurthy, Vijay, d'Aspremont, Alexandre
#FOS: Mathematics #Optimization and Control (math.OC)

paper · doi:10.48550/arxiv.0908.0143

Abstract

Covariance selection seeks to estimate a covariance matrix by maximum likelihood while restricting the number of nonzero inverse covariance matrix coefficients. A single penalty parameter usually controls the tradeoff between log likelihood and sparsity in the inverse matrix. We describe an efficient algorithm for computing a full regularization path of solutions to this problem.

Related

• First-order methods for sparse covariance selection
• Sparse Covariance Selection via Robust Maximum Likelihood Estimation
• Towards a sparse, scalable, and stably positive definite (inverse) covariance estimator
• Adaptive First-Order Methods for General Sparse Inverse Covariance Selection
• A path following algorithm for Sparse Pseudo-Likelihood Inverse Covariance Estimation (SPLICE)