BibTeX
@MISC{Populations_howeffectiveare,
author = {Multiple Populations and William F. Punch},
title = {Howeffective Are},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The use of multiple populations in Genetic Programming is an area that is just beginning to be investigated. To date a number of conflicting reports have been generated with respect to the effectiveness of multiple populations in GP. We report here that these conflicting reports may be due a problem-dependent nature found in GP that has not been reported in GAs. This paper will review: what both multiple populations and speed-up mean in the areas of GA/GP, some conflicting results that have been reported in the GP literature on whether multiple populations gives speed-up to GP problems, and offer an answer as to why different GP problems show different kinds of speed-up when using multiple populations. 1.0 INTRODUCTION Parallel processing is an issue that has often been examined in Genetic Algorithm (GA) research [Lin 1994, Manderick and Spiessens 1989, Mulhenbein 1989, Punch 1993, Tanese 1989, Pettey 1987]. It is worth noting that there are at least three senses of parallelism in th...
