Approximate Message Passing Algorithm for Nonconvex Regularization
Open Access
- 7 January 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 9080-9090
Abstract
In this paper, we study the sparse signal reconstruction with nonconvex regularization, mainly focus on two popular nonconvex regularization methods, minimax concave penalty (MCP) and smoothly clipped absolute deviation (SCAD). Approximate message passing (AMP) algorithm is an effective method for signal reconstruction. Based on AMP algorithm, we propose an improved MCP iterative thresholding algorithm and an improved SCAD iterative thresholding algorithm. Further, we analyze the convergence of the new algorithms, and provide a series of experiments to assess the performance of the new algorithms. The experiments show that the new algorithms based on AMP have stronger reconstruction capabilities, higher phase transition for sparse signal reconstruction and better variable selection ability than original MCP iterative thresholding algorithm and original SCAD iterative thresholding algorithm.Keywords
Funding Information
- Macau Science and Technology Development Funds from the Macau Special Administrative Region, China (003/2016/AFJ)
- National Grand Fundamental Research 973 Program of China (2013CB329404)
- National Natural Science Foundation of China (61373114, 61661166011, 11690011, 61721002)
- National Natural Science Foundation of China (11571011)
This publication has 24 references indexed in Scilit:
- $L_{1/2}$ Regularization: A Thresholding Representation Theory and a Fast SolverIEEE Transactions on Neural Networks and Learning Systems, 2012
- Statistical-Physics-Based Reconstruction in Compressed SensingPhysical Review X, 2012
- Nearly unbiased variable selection under minimax concave penaltyThe Annals of Statistics, 2010
- Message-passing algorithms for compressed sensingProceedings of the National Academy of Sciences, 2009
- The Dantzig selector: Statistical estimation when p is much larger than nThe Annals of Statistics, 2007
- Graphical ModelsStatistical Science, 2004
- Uncertainty principles and ideal atomic decompositionIEEE Transactions on Information Theory, 2001
- Atomic Decomposition by Basis PursuitSIAM Review, 2001
- Sparse Approximate Solutions to Linear SystemsSIAM Journal on Computing, 1995
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978