English

Gradient Support Projection Algorithm for Affine Feasibility Problem with Sparsity and Nonnegativity

Optimization and Control 2014-06-30 v1

Abstract

Let AA be a real M×NM \times N measurement matrix and bRMb\in \mathbb{R}^M be an observations vector. The affine feasibility problem with sparsity and nonnegativity (AFPSNAFP_{SN} for short) is to find a sparse and nonnegative vector xRNx\in \mathbb{R}^N with Ax=bAx=b if such xx exists. In this paper, we focus on establishment of optimization approach to solving the AFPSNAFP_{SN}. By discussing tangent cone and normal cone of sparse constraint, we give the first necessary optimality conditions, α\alpha-Stability, T-Stability and N-Stability, and the second necessary and sufficient optimality conditions for the related minimization problems with the AFPSNAFP_{SN}. By adopting Armijo-type stepsize rule, we present a framework of gradient support projection algorithm for the AFPSNAFP_{SN} and prove its full convergence when matrix AA is ss-regular. By doing some numerical experiments, we show the excellent performance of the new algorithm for the AFPSNAFP_{SN} without and with noise.

Keywords

Cite

@article{arxiv.1406.7178,
  title  = {Gradient Support Projection Algorithm for Affine Feasibility Problem with Sparsity and Nonnegativity},
  author = {Lili Pan and Naihua Xiu and Shenglong Zhou},
  journal= {arXiv preprint arXiv:1406.7178},
  year   = {2014}
}
R2 v1 2026-06-22T04:49:17.177Z