Sparse regularization with the ℓ0 norm
From MaRDI portal
Abstract: We consider a minimization problem whose objective function is the sum of a fidelity term, not necessarily convex, and a regularization term defined by a positive regularization parameter multiple of the norm composed with a linear transform. This problem has wide applications in compressed sensing, sparse machine learning and image reconstruction. The goal of this paper is to understand what choices of the regularization parameter can dictate the level of sparsity under the transform for a global minimizer of the resulting regularized objective function. This is a critical issue but it has been left unaddressed. We address it from a geometric viewpoint with which the sparsity partition of the image space of the transform is introduced. Choices of the regularization parameter are specified to ensure that a global minimizer of the corresponding regularized objective function achieves a prescribed level of sparsity under the transform. Results are obtained for the spacial sparsity case in which the transform is the identity map, a case that covers several applications of practical importance, including machine learning, image/signal processing and medical image reconstruction.
Recommendations
- Description of the minimizers of least squares regularized with _0-norm. Uniqueness of the global minimizer
- Well-posedness classes for sparse regularization
- Hard thresholding pursuit with continuation for \(\ell^{0}\)-regularized minimizations
- Entropic regularization of the \(\ell _{0}\) function
- Capped \(\ell_p\) approximations for the composite \(\ell_0\) regularization problem
Cites work
- A representer theorem for deep neural networks
- Affine systems in \(L_ 2(\mathbb{R}^d)\): The analysis of the analysis operator
- Aggregation of regularized solutions from multiple observation models
- An approximate sparsity model for inpainting
- Convex analysis and monotone operator theory in Hilbert spaces
- Denoising point sets via \(L_0\) minimization
- Fast collocation methods for solving ill-posed integral equations of the first kind
- Filters of wavelets on invariant sets for image denoising
- High dimensional thresholded regression and shrinkage effect
- scientific article; zbMATH DE number 1465030 (Why is no real title available?)
- scientific article; zbMATH DE number 6982301 (Why is no real title available?)
- Inverting incomplete Fourier transforms by a sparse regularization model and applications in seismic wavefield modeling
- Linear integral equations
- Matrix completion via minimizing an approximate rank
- Mesh denoising via L 0 minimization
- Multi-parameter Tikhonov regularization with the ^0 sparsity constraint
- Multiscale methods for Fredholm integral equations
- Noisy 1-bit compressive sensing: models and algorithms
- Nonlinear total variation based noise removal algorithms
- Reconstruction and decomposition algorithms for biorthogonal multiwavelets
- Robust Linear Regression via \ell₀ Regularization
- Sparsity promoting regularization for effective noise suppression in SPECT image reconstruction
- Ten Lectures on Wavelets
- The Discrete Cosine Transform
- Tight frame: an efficient way for high-resolution image reconstruction
- Tikhonov regularization with \({\ell^{0}}\)-term complementing a convex penalty: \({\ell^{1}}\)-convergence under sparsity constraints
- Tikhonov regularization with oversmoothing penalty for non-linear ill-posed problems in Hilbert scales
- Using the matrix refinement equation for the construction of wavelets on invariant sets
- Wavelet inpainting with the \(\ell_{0}\) sparse regularization
Cited in
(20)- Average performance of the approximation in a dictionary using an \(\ell _0\) objective
- Capped \(\ell_p\) approximations for the composite \(\ell_0\) regularization problem
- Sparsity-enforcing regularisation and ISTA revisited
- Relationship between the optimal solutions of least squares regularized with \(\ell_{0}\)-norm and constrained by \(k\)-sparsity
- Description of the minimizers of least squares regularized with _0-norm. Uniqueness of the global minimizer
- Entropic regularization of the \(\ell _{0}\) function
- A regularized sparse representation method
- Norm sensitivity of sparsity regularization with respect to p
- A continuous exact _0 penalty (CEL0) for least squares regularized problem
- Sparse Regularization via Convex Analysis
- Parameter choices for sparse regularization with the ℓ1 norm *
- scientific article; zbMATH DE number 6302955 (Why is no real title available?)
- Neural network for a class of sparse optimization with L₀-regularization
- Inexact fixed-point proximity algorithm for the _0 sparse regularization problem
- Sparsity-guided multi-parameter selection in _1-regularized models via a fixed-point proximity approach
- New conditions on robust and stable recovery of non-strictly k-ranked matrices via Schatten p-norm minimization
- A gradient-based learning method with smoothing group L₀ regularization for interval perceptron and interval weights
- MG-SpaIR: multi-grade sparse-guided implicit representation for training-data-free image restoration
- Operator sequence-based phase retrieval on quaternionic left Hilbert spaces
- Minimizers of sparsity regularized Huber loss function
This page was built for publication: Sparse regularization with the ℓ0 norm
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6166163)
