A method for finding structured sparse solutions to nonnegative least squares problems with applications
From MaRDI portal
(Redirected from Publication:2873273)
Abstract: Demixing problems in many areas such as hyperspectral imaging and differential optical absorption spectroscopy (DOAS) often require finding sparse nonnegative linear combinations of dictionary elements that match observed data. We show how aspects of these problems, such as misalignment of DOAS references and uncertainty in hyperspectral endmembers, can be modeled by expanding the dictionary with grouped elements and imposing a structured sparsity assumption that the combinations within each group should be sparse or even 1-sparse. If the dictionary is highly coherent, it is difficult to obtain good solutions using convex or greedy methods, such as non-negative least squares (NNLS) or orthogonal matching pursuit. We use penalties related to the Hoyer measure, which is the ratio of the and norms, as sparsity penalties to be added to the objective in NNLS-type models. For solving the resulting nonconvex models, we propose a scaled gradient projection algorithm that requires solving a sequence of strongly convex quadratic programs. We discuss its close connections to convex splitting methods and difference of convex programming. We also present promising numerical results for example DOAS analysis and hyperspectral demixing problems.
Recommendations
- Provably optimal sparse solutions to overdetermined linear systems with non-negativity constraints in a least-squares sense by implicit enumeration
- Active set type algorithms for nonnegative matrix factorization in hyperspectral unmixing
- Fast nonnegative least squares through flexible Krylov subspaces
- Inverse problems with nonnegative and sparse solutions: algorithms and application to the phase retrieval problem
- Nonnegative Matrix Factorization Based on Alternating Nonnegativity Constrained Least Squares and Active Set Method
Cited in
(68)- Three \(l_1\) based nonconvex methods in constructing sparse mean reverting portfolios
- DC programming and DCA for solving Brugnano-Casulli piecewise linear systems
- \(l_1\)-\(l_2\) regularization of split feasibility problems
- Fast L1-L2 minimization via a proximal operator
- DC programming and DCA: thirty years of developments
- Minimization of transformed L₁ penalty: theory, difference of convex function algorithm, and robust application in compressed sensing
- Solving structured nonsmooth convex optimization with complexity \(\mathcal {O}(\varepsilon ^{-1/2})\)
- Analysis of the ratio of \(\ell_1\) and \(\ell_2\) norms in compressed sensing
- High-dimensional sign-constrained feature selection and grouping
- Provably optimal sparse solutions to overdetermined linear systems with non-negativity constraints in a least-squares sense by implicit enumeration
- Penalized robust estimators in sparse logistic regression
- The proximity operator of the log-sum penalty
- Unconstrained \(\ell_1\)-\(\ell_2\) minimization for sparse recovery via mutual coherence
- The springback penalty for robust signal recovery
- Smoothing inertial projection neural network for minimization \(L_{p-q}\) in sparse signal reconstruction
- Transformed \(\ell_1\) regularization for learning sparse deep neural networks
- Sparse signal reconstruction via the approximations of \(\ell_0\) quasinorm
- A class of null space conditions for sparse recovery via nonconvex, non-separable minimizations
- Reconstruction of jointly sparse vectors via manifold optimization
- An optimal subgradient algorithm for large-scale bound-constrained convex optimization
- A unified DC programming framework and efficient DCA based approaches for large scale batch reinforcement learning
- A necessary and sufficient condition for sparse vector recovery via \(\ell_1-\ell_2\) minimization
- Efficient color image segmentation via quaternion-based \(L_1/L_2\) Regularization
- Morozov's discrepancy principle for _1-_2 sparsity regularization
- Truncated $l_{1-2}$ Models for Sparse Recovery and Rank Minimization
- A weighted difference of anisotropic and isotropic total variation model for image processing
- Heuristic discrepancy principle for variational regularization of inverse problems
- Iterative positive thresholding algorithm for non-negative sparse optimization
- Multicompartment magnetic resonance fingerprinting
- Total variation regularization strategies in full-waveform inversion
- Computing sparse representation in a highly coherent dictionary based on difference of L₁ and L₂
- _1-_2 minimization methods for signal and image reconstruction with impulsive noise removal
- A three-operator splitting algorithm for nonconvex sparsity regularization
- A new sufficient condition for sparse vector recovery via _1- _2 local minimization
- Minimization of L₁ over L₂ for sparse signal recovery with convergence guarantee
- An iterative reduction FISTA algorithm for large-scale LASSO
- A Scale-Invariant Approach for Sparse Signal Recovery
- \(\alpha\ell_1-\beta\ell_2\) regularization for sparse recovery
- An efficient proximal block coordinate homotopy method for large-scale sparse least squares problems
- Minimization of \(\ell_{1-2}\) for compressed sensing
- Sparse solution of nonnegative least squares problems with applications in the construction of probabilistic Boolean networks.
- A smoothing inertial neural network for sparse signal reconstruction with noise measurements via L_p-L₁ minimization
- Block-sparse recovery and rank minimization using a weighted \(l_p-l_q\) model
- Matrix-wise _0-constrained sparse nonnegative least squares
- Study on \(L_1\) over \(L_2\) Minimization for nonnegative signal recovery
- Structured model selection via ℓ1−ℓ2 optimization
- Nonconvex _p-_q minimization method and p-RIP condition for stable recovery of approximately k-sparse signals
- A reduced half thresholding algorithm
- The Lawson‐Hanson algorithm with deviation maximization: Finite convergence and sparse recovery
- Open issues and recent advances in DC programming and DCA
- An iDCA with sieving strategy for PDE-constrained optimization problems with \(L^{1-2}\)-control cost
- A parameterized three-operator splitting algorithm for non-convex minimization problems with applications
- Sparse recovery under nonnegativity and sum-to-one constraints
- Sparse recovery with coherent frames via \(\ell_{1-2}\)-analysis
- Subspace Newton method for sparse group \(\ell_0\) optimization problem
- From theoretical guarantee to practical performance: selectable and optimal step-lengths for IHT and HTP algorithms in compressed sensing
- A nonlocal weighted difference of anisotropic and isotropic total variation to regularize partition boundaries in an image
- Tensor completion by L_-L_F bilevel programming
- Parameterized proximal-gradient algorithms for L1/L2 sparse signal recovery
- Identification of corrupted data via k-means clustering for function approximation
- An extended sequential quadratic method with extrapolation
- Sparse portfolio optimization via _1 over _2 regularization
- Deep image prior and weighted anisotropic-isotropic total variation regularization for solving linear inverse problems
- Iterative mix thresholding algorithm with continuation technique for mix sparse optimization and application
- A general framework for group sparsity in hyperspectral unmixing using endmember bundles
- On the hardness of the L₁-L₂ regularization problem
- An efficient proximal algorithm for squared L1 over L2 regularized sparse recovery
- Recovery bounds for cardinality regularized optimization problem
This page was built for publication: A method for finding structured sparse solutions to nonnegative least squares problems with applications
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2873273)
