Computational and Applied Math Proseminar

Wednesday, March 5, 1:40 p.m. GWC 487

Yin Zhang

Dept Comp Appl Math, Rice U

Large-Scale L1-Related Minimization
in Compressive Sensing and Beyond

Abstract
Compressive sensing (CS) has recently been actively studied because of its promises in many potential applications. From CS and other related areas arise a number of large-scale optimization problems with non-smooth regularization terms such as L1-norm.

In this talk, we will consider several such formulations, discuss algorithmic challenges involved, and introduce some existing approaches. In particular, we will focus on theoretical and numerical results for a fixed-point continuation (FPC) algorithm. We will also briefly discuss issues of exploiting problem structures.

For further information please contact: mittelmann@asu.edu