Probabilistic analysis of the phase space flow for linear programming
Abstract
The phase space flow of a dynamical system leading to the solution of Linear Programming (LP) problems is explored as an example of complexity analysis in an analog computation framework. An ensemble of LP problems with variables and constraints (), where all elements of the vectors and matrices are normally distributed is studied. The convergence time of a flow to the fixed point representing the optimal solution is computed. The cumulative distribution of the convergence rate to this point is calculated analytically, in the asymptotic limit of large , in the framework of Random Matrix Theory. In this limit is found to be a scaling function, namely it is a function of one variable that is a combination of , and rather then a function of these three variables separately. From numerical simulations also the distribution of the computation times is calculated and found to be a scaling function as well.
Cite
@article{arxiv.cond-mat/0110655,
title = {Probabilistic analysis of the phase space flow for linear programming},
author = {Asa Ben-Hur and Joshua Feinberg and Shmuel Fishman and Hava T. Siegelmann},
journal= {arXiv preprint arXiv:cond-mat/0110655},
year = {2009}
}
Comments
8 pages, latex, 2 eps figures; final published version