A FIRST-ORDER SMOOTHED PENALTY METHOD FOR COMPRESSED SENSING

被引:23
作者
Aybat, N. S. [1 ]
Iyengar, G. [1 ]
机构
[1] Columbia Univ, Dept Ind Engn & Operat Res, New York, NY 10027 USA
基金
美国国家科学基金会;
关键词
l(1)-minimization; basis pursuit; compressed sensing; sparse optimization; smooth approximation of nonsmooth functions; Nesterov's method; ROBUST UNCERTAINTY PRINCIPLES; LINEAR INVERSE PROBLEMS; RECONSTRUCTION; L(1)-MINIMIZATION; CONTINUATION; ALGORITHM;
D O I
10.1137/090762294
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
We propose a first-order smoothed penalty algorithm (SPA) to solve the sparse recovery problem min{parallel to x parallel to(1) : Ax = b}. SPA is efficient as long as the matrix-vector product Ax and A(T) y can be computed efficiently; in particular, A need not have orthogonal rows. SPA converges to the target signal by solving a sequence of penalized optimization subproblems, and each subproblem is solved using Nesterov's optimal algorithm for simple sets [Yu. Nesterov, Introductory Lectures on Convex Optimization: A Basic Course, Kluwer Academic Publishers, Norwell, MA, 2004] and [Yu. Nesterov, Math. Program., 103 (2005), pp. 127-152]. We show that the SPA iterates x(k) are epsilon-feasible; i.e. parallel to Ax(k) - b parallel to(2) <= epsilon and epsilon-optimal; i.e. vertical bar parallel to x(psi)parallel to 1 vertical bar <= epsilon after (O) over tilde (epsilon(-3/2)) iterations. SPA is able to work with l(1), l(2), or l(infinity) penalty on the infeasibility, and SPA can be easily extended to solve the relaxed recovery problem min{parallel to x parallel to(1) : parallel to Ax - b parallel to(2) <= delta}.
引用
收藏
页码:287 / 313
页数:27
相关论文
共 20 条
[11]  
Hale E. T., 2007, TR07 RIC U
[12]   FIXED-POINT CONTINUATION FOR l1-MINIMIZATION: METHODOLOGY AND CONVERGENCE [J].
Hale, Elaine T. ;
Yin, Wotao ;
Zhang, Yin .
SIAM JOURNAL ON OPTIMIZATION, 2008, 19 (03) :1107-1130
[13]   Smoothing Techniques for Computing Nash Equilibria of Sequential Games [J].
Hoda, Samid ;
Gilpin, Andrew ;
Pena, Javier ;
Sandholm, Tuomas .
MATHEMATICS OF OPERATIONS RESEARCH, 2010, 35 (02) :494-512
[14]  
Huber P., 2011, ROBUST STAT
[15]  
Koh K., 2007, Solver for l1-regularized least squares problems
[16]   Smooth minimization of non-smooth functions [J].
Nesterov, Y .
MATHEMATICAL PROGRAMMING, 2005, 103 (01) :127-152
[17]  
Nesterov Y., 2004, INTRO LECTURES CONVE
[18]   PROBING THE PARETO FRONTIER FOR BASIS PURSUIT SOLUTIONS [J].
van den Berg, Ewout ;
Friedlander, Michael P. .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2008, 31 (02) :890-912
[19]   A FAST ALGORITHM FOR SPARSE RECONSTRUCTION BASED ON SHRINKAGE, SUBSPACE OPTIMIZATION, AND CONTINUATION [J].
Wen, Zaiwen ;
Yin, Wotao ;
Goldfarb, Donald ;
Zhang, Yin .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2010, 32 (04) :1832-1857
[20]   Bregman Iterative Algorithms for l1-Minimization with Applications to Compressed Sensing [J].
Yin, Wotao ;
Osher, Stanley ;
Goldfarb, Donald ;
Darbon, Jerome .
SIAM JOURNAL ON IMAGING SCIENCES, 2008, 1 (01) :143-168