Abstract:Sparse recovery has attracted considerable attention in signal processing community these years, because of its widespread usage in many applications. Though lots of convex and nonconvex methods have been proposed for this topic, most of them are computationally expensive. Recently, learned iterative shrinkage-thresholding algorithm (LISTA) and its variants are incorporated to sparse recovery, which are based on specific deep neural networks designed by unfolding ISTA. However, the problem they are able to solve is regularized by 11 norm, which is less effective at promoting sparseness than some nonconvex sparsity-inducing penalties (SIPs). Motivated by the fact that the problems regularized by these SIPs can be solved by proximal operator methods, we devise a network named learning proximal operator method (LePOM) for sparse recovery. For LePOM on a general class of SIPs, a necessary condition of its convergence is established. Based on this condition, a simplified network is proposed with less parameters. Theoretical analysis of this network inspires us to further reduce the number of parameters and arrive at proposing Analytical LePOM (ALePOM). ALePOM determines most parameters by solving an optimization problem and significantly reduces the number of parameters. Theoretical analysis shows if the signal is sufficiently sparse, ALePOM converges linearly. Simulations confirm our analyses and demonstrate that the proposed solutions outperform state-of-the-art sparse recovery algorithms and neural network-based methods.

Efficient Sparse Recovery Via Adaptive Non-Convex Regularizers with Oracle Property

A Non-convex Approach for Sparse Recovery with Convergence Guarantee

Efficient and Robust Recovery of Sparse Signal and Image Using Generalized Nonconvex Regularization

Robust Sparse Recovery Via Non-Convex Optimization

Sparse Recovery Via Nonconvex Regularized M-estimators over ℓ_q-Balls

The Convergence Guarantees of a Non-Convex Approach for Sparse Recovery.

Efficient Sparse Recovery and Demixing Using Nonconvex Regularization

Fast Sparse Recovery Via Non-Convex Optimization

A Fast Convex Conjugated Algorithm for Sparse Recovery

Nonconvex relaxation approaches to robust matrix recovery

Improving Spiking Sparse Recovery via Non-Convex Penalties

Low-rank matrix recovery with non-quadratic loss: projected gradient method and regularity projection oracle

A Robust Recovery Algorithm with Smoothing Strategies.

An Efficient Semismooth Newton Method for Adaptive Sparse Signal Recovery Problems

A Unified Primal Dual Active Set Algorithm for Nonconvex Sparse Recovery

A Survey on Nonconvex Regularization Based Sparse and Low-Rank Recovery in Signal Processing, Statistics, and Machine Learning

Learning Proximal Operator Methods for Nonconvex Sparse Recovery with Theoretical Guarantee.

A General Iterative Shrinkage and Thresholding Algorithm for Non-convex Regularized Optimization Problems

Robust Sparse Recovery Via Weakly Convex Optimization in Impulsive Noise

Convex Relaxation Algorithm for a Structured Simultaneous Low-Rank and Sparse Recovery Problem

A Sparse Recovery Algorithm Based on Arithmetic Optimization