Sparse Regularization via Convex Analysis

Abstract

Sparse approximate solutions to linear equations are classically obtained via L1 norm regularized least squares, but this method often underestimates the true solution. As an alternative to the L1 norm, this paper proposes a class of non-convex penalty functions that maintain the convexity of the least squares cost function to be minimized, and avoids the systematic underestimation characteristic of L1 norm regularization. The proposed penalty function is a multivariate generalization of the minimax-concave (MC) penalty. It is defined in terms of a new multivariate generalization of the Huber function, which in turn is defined via infimal convolution. The proposed sparse-regularized least squares cost function can be minimized by proximal algorithms comprising simple computations.

References in corpus (18)

• Nearly unbiased variable selection under minimax concave penalty
• A fast approach for overcomplete sparse decomposition based on smoothed L0 norm
• The Convex Geometry of Linear Inverse Problems
• One-step sparse estimates in nonconcave penalized likelihood models
• The Cosparse Analysis Model and Algorithms
• Group-Sparse Signal Denoising: Non-Convex Regularization, Convex Optimization
• A proximal iteration for deconvolving Poisson noisy images using sparse representations
• Sparse Signal Estimation by Maximally Sparse Convex Optimization
• Compressed Sensing Recovery via Nonconvex Shrinkage Penalties
• Sparsity-based Algorithm for Detecting Faults in Rotating Machines
• A Majorize-Minimize subspace approach for l2-l0 image regularization
• Total Variation Denoising via the Moreau Envelope
• Enhanced Low-Rank Matrix Approximation
• On the Convergence of the Iterative Shrinkage/Thresholding Algorithm With a Weakly Convex Penalty
• Enhanced Sparsity by Non-Separable Regularization
• Homotopy based algorithms for ${\ell }_0$-regularized least-squares
• A Class of Nonconvex Penalties Preserving Overall Convexity in Optimization-Based Mean Filtering
• On convexification/optimization of functionals including an l2-misfit term

Cited by in corpus (28)

• ${\ell }_0$-Motivated Low-Rank Sparse Subspace Clustering
• Ship Wake Detection in SAR Images via Sparse Regularization
• On Solving SAR Imaging Inverse Problems Using Non-Convex Regularization with a Cauchy-based Penalty
• Convergence Guarantees for Non-Convex Optimisation with Cauchy-Based Penalties
• Linearly Involved Generalized Moreau Enhanced Models and Their Proximal Splitting Algorithm under Overall Convexity Condition
• Generalized L${}_p$-norm joint inversion of gravity and magnetic data using cross-gradient constraint
• Nonconvex penalties with analytical solutions for one-bit compressive sensing
• Image Fusion via Sparse Regularization with Non-Convex Penalties
• Linearly-involved Moreau-Enhanced-over-Subspace Model: Debiased Sparse Modeling and Stable Outlier-Robust Regression
• An unbiased approach to compressed sensing
• A Unified Framework for Solving a General Class of Nonconvexly Regularized Convex Models
• Detection of Ship Wakes in SAR Imagery Using Cauchy Regularisation
• Sparse Signal Reconstruction for Nonlinear Models via Piecewise Rational Optimization
• Monotone Lipschitz-Gradient Denoiser: Explainability of Operator Regularization Approaches Free From Lipschitz Constant Control
• DuRIN: A Deep-unfolded Sparse Seismic Reflectivity Inversion Network
• A Convex-Nonconvex Strategy for Grouped Variable Selection
• Continuous Relaxation of Discontinuous Shrinkage Operator: Proximal Inclusion and Conversion
• Unsupervised Multi-view Clustering by Squeezing Hybrid Knowledge from Cross View and Each View
• A Survey on Nonconvex Regularization Based Sparse and Low-Rank Recovery in Signal Processing, Statistics, and Machine Learning
• NuSPAN: A Proximal Average Network for Nonuniform Sparse Model -- Application to Seismic Reflectivity Inversion
• Exploiting the Dual-Tree Complex Wavelet Transform for Ship Wake Detection in SAR Imagery
• A Convex-Nonconvex Framework for Enhancing Minimization Induced Penalties
• SWAGGER: Sparsity Within and Across Groups for General Estimation and Recovery
• Sparse Optimization Problem with s-difference Regularization
• Non-convex optimization via strongly convex majoirziation-minimization
• A Convexly Constrained LiGME Model and Its Proximal Splitting Algorithm
• RINS-T: Robust Implicit Neural Solvers for Time Series Linear Inverse Problems
• A Sparsity Algorithm with Applications to Corporate Credit Rating