Abstract:We consider the covariance selection problem where variables are clustered into groups and the inverse covariance matrix is expected to have a blockwise sparse structure. This problem is realized via penalizing the maximum likelihood estimation of the inverse covariance matrix by group Lasso regularization. We propose to solve the resulting log-determinant optimization problem with the classical proximal point algorithm (PPA). At each iteration, as it is difficult to update the primal variables directly, we first solve the dual subproblem by an inexact semismooth Newton-CG method and then update the primal variables by explicit formulas based on the computed dual variables. We also propose to accelerate the PPA by an inexact generalized Newton's method when the iterate is close to the solution. Theoretically, we prove that at the optimal solution, the nonsingularity of the generalized Hessian matrices of the dual subproblem is equivalent to the constraint nondegeneracy condition for the primal problem. Global and local convergence results are also presented for the proposed PPA. Moreover, based on the augmented Lagrangian function of the dual problem we derive an alternating direction method (ADM), which is easily implementable and is demonstrated to be efficient for random problems. Numerical results, including comparisons with the ADM on both synthetic and real data, are presented to demonstrate that the proposed Newton-CG based PPA is stable and efficient and, in particular, outperforms the ADM when high accuracy is required.

Robust local K-proximal plane clustering based on L2,1-norm minimization

Robust Principal Component Analysis Based on Maximum Correntropy Criterion

L1-Norm Distance Minimization-Based Fast Robust Twin Support Vector $k$ -Plane Clustering.

Subspace clustering via joint l1,2 and l2,1 norms

Clustering to Minimize Cluster-Aware Norm Objectives

Provably noise-robust, regularised $k$-means clustering

Avoiding Optimal Mean Robust PCA/2DPCA with Non-Greedy L1-norm Maximization

Robust Principal Component Analysis Via Joint ℓ<inf>2,1</inf>-Norms Minimization

Avoiding Optimal Mean ℓ 2,1 -Norm Maximization-Based Robust PCA for Reconstruction

Non-Greedy L21-Norm Maximization for Principal Component Analysis

Robust and Efficient Fuzzy C-Means Clustering Constrained on Flexible Sparsity

Robust principal component analysis via optimal mean by joint ℓ

A Proximal Point Algorithm For Log-Determinant Optimization With Group Lasso Regularization

Probabilistic K-means Clustering via Nonlinear Programming

Robust Principal Component Analysis via Joint l(2,1)-Norms Minimization

Robust Principal Component Analysis Via Optimal Mean by Joint ℓ2,1 and Schatten P-Norms Minimization

Robust DLPP With Nongreedy ℓ1-Norm Minimization and Maximization.

Robust Subspace Clustering via Tighter Rank Approximation

Point Cloud Normal Estimation Via Low-Rank Subspace Clustering

Robust Subspace Clustering via Smoothed Rank Approximation

A Riemannian Optimization Approach to Clustering Problems