BibTeX
@MISC{Hao_anexperimental,
author = {Jin-Kao Hao and Jérôme Pannier},
title = {An Experimental Study of Local Search for Constraint Solving},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
In this paper, we present an experimental study of local search for constraint satisfaction. For this purpose, we experiment with two algorithms based on Simulated Annealing (SA) and Tabu Search (TS) for solving the maximal constraint satisfaction problem. These two algorithms were tested on various large instances going from 100 to 300 variables with 15 to 30 values per variable. Experimental results show that the TS algorithm dominates the RS algorithm on all the tested instances in terms of solution quality and solving speed. We propose empirical arguments to explain the difference of performances.
