Geometric Presolver example

From Wikimization

Revision as of 16:54, 11 April 2013 by Dattorro (Talk | contribs)
Jump to: navigation, search

Assume that the following problem is massive:

LaTeX: \begin{array}{rl}\mbox{find}&x\\
\mbox{subject to}&E\,x=t\\
&x\succeq_{}\mathbf{0}\end{array}

The problem is presumed solvable but not computable by any contemporary means. The most logical strategy is to make the problem smaller.

This Matlab workspace file contains a real LaTeX: E matrix having dimension LaTeX: 533\times 2704 and compatible LaTeX: t vector. There exists a cardinality LaTeX: 36 binary solution LaTeX: x. Before attempting to find it, we presume to have no choice but to reduce dimension of the LaTeX: E matrix prior to computing a solution.

A lower bound on the number of rows of LaTeX: \,E\in\mathbb{R}^{533\times 2704}\, retained is LaTeX: 217.
A lower bound on the number of columns retained is LaTeX: 1104.

The present exercise is to determine those rows and columns using any contemporary presolver.

Personal tools