Sparse Regularization via Convex Analysis

DOI10.1109/TSP.2017.2711501zbMATH Open1414.94545arXiv1803.06765OpenAlexW2620850570MaRDI QIDQ4621829FDOQ4621829

Authors: Ivan W. Selesnick 

Publication date: 8 February 2019

Published in: IEEE Transactions on Signal Processing (Search for Journal in Brave)

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.

Full work available at URL: https://arxiv.org/abs/1803.06765

Recommendations

• Sparse Convex Regression
• Analysis of multi-stage convex relaxation for sparse regularization
• Non-convex sparse regularisation
• Robust Sparse Analysis Regularization
• Nonconvex sparse regularization and splitting algorithms
• The Geometry of Sparse Analysis Regularization
• Structured sparsity through convex optimization
• Regularizers for structured sparsity
• Sparse regularization with the ℓ0 norm
• Nonparametric sparsity and regularization

Mathematics Subject Classification ID

Convex programming (90C25) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Cited In (35)

• A characterization of proximity operators
• Two-stage convex relaxation approach to low-rank and sparsity regularized least squares loss
• An unbiased approach to compressed sensing
• Minimizers of sparsity regularized Huber loss function
• Robust nonconvex sparse optimization for impact force identification
• Non-convex optimization via strongly convex majorization-minimization
• Sparse Approximation via Penalty Decomposition Methods
• Sparsity-Inducing Nonconvex Nonseparable Regularization for Convex Image Processing
• Low-CP-rank tensor completion via practical regularization
• Weakly decomposable regularization penalties and structured sparsity
• Regularity properties for sparse regression
• Alternating direction method of multipliers for nonconvex log total variation image restoration
• Tight-frame-like analysis-sparse recovery using nontight sensing matrices
• Sparse solutions to an underdetermined system of linear equations via penalized Huber loss
• Convex non-convex variational models
• New insights on the optimality conditions of the \(\ell_2-\ell_0\) minimization problem
• Diffusion optical tomography reconstruction based on convex–nonconvex graph total variation regularization
• A continuous relaxation of the constrained \(\ell_2-\ell_0\) problem
• Sparsity-enforcing regularisation and ISTA revisited
• A convex-Nonconvex strategy for grouped variable selection
• Penalized least square in sparse setting with convex penalty and non Gaussian errors
• Necessary and Sufficient Conditions for Noiseless Sparse Recovery via Convex Quadratic Splines
• Norm sensitivity of sparsity regularization with respect to p
• Bregman distance regularization for nonsmooth and nonconvex optimization
• A solution approach for cardinality minimization problem based on fractional programming
• Sparse reconstruction via the mixture optimization model with iterative support estimate
• Inverse problems are solvable on real number signal processing hardware
• Plug-and-play algorithms for convex non-convex regularization: convergence analysis and applications
• Parametrized quasi-soft thresholding operator for compressed sensing and matrix completion
• Sparse signal reconstruction via the approximations of \(\ell_0\) quasinorm
• Sparse and robust estimation with ridge minimax concave penalty
• A new hybrid \(l_p\)-\(l_2\) model for sparse solutions with applications to image processing
• The Geometry of Sparse Analysis Regularization
• Linearly involved generalized Moreau enhanced models and their proximal splitting algorithm under overall convexity condition
• Non-convex total variation regularization for convex denoising of signals

This page was built for publication: Sparse Regularization via Convex Analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4621829)