A first-order smoothed penalty method for compressed sensing

N. S. Aybat, G. Iyengar

Research output: Contribution to journalArticle

15 Scopus citations

Abstract

We propose a first-order smoothed penalty algorithm (SPA) to solve the sparse recovery problem min{||x||1: Ax = b}. SPA is efficient as long as the matrix-vector product Ax and AT 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 xk are ∈-feasible; i.e. || Axk - b ||2 ≤ e and e-optimal; i.e.

Original languageEnglish (US)
Pages (from-to)287-313
Number of pages27
JournalSIAM Journal on Optimization
Volume21
Issue number1
DOIs
StatePublished - May 30 2011

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science

Fingerprint Dive into the research topics of 'A first-order smoothed penalty method for compressed sensing'. Together they form a unique fingerprint.

Cite this