Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Dynamic-probabilistic particle swarms

Dynamic-probabilistic particle swarms Dynamic-Probabilistic Particle Swarms James Kennedy US Bureau of Labor Statistics Washington DC 20212 Kennedy_Jim@bls.gov ABSTRACT The particle swarm algorithm is usually a dynamic process, where a point in the search space to be tested depends on the previous point and the direction of movement. The process can be decomposed, and probability distributions around a center can be used instead of the usual trajectory approach. A version that is both dynamic and Gaussian looks very promising. ACM Categories & Subject Descriptors I.2.11 Distributed Artificial Intelligence Multiagent systems Keywords: Particle swarms 1. INTRODUCTION Since its introduction in 1995 (Kennedy and Eberhart, 1995; Eberhart and Kennedy, 1995), the particle swarm algorithm has gone through many changes. Though early results were surprisingly good, and though the method had very few moving parts, it turned out to be quite difficult to understand how it worked, in order to improve it. Over the past decade, numerous modifications have been introduced, several of which have turned out to cause genuine improvements in performance, and several of which have helped to understand the dynamics of the swarm and how it is able to solve problems. The canonical particle swarm algorithm is given as: For each population http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Dynamic-probabilistic particle swarms

Association for Computing Machinery — Jun 25, 2005

Loading next page...
/lp/association-for-computing-machinery/dynamic-probabilistic-particle-swarms-NuzYiSF8zb

References (13)

Datasource
Association for Computing Machinery
Copyright
Copyright © 2005 by ACM Inc.
ISBN
1-59593-010-8
doi
10.1145/1068009.1068040
Publisher site
See Article on Publisher Site

Abstract

Dynamic-Probabilistic Particle Swarms James Kennedy US Bureau of Labor Statistics Washington DC 20212 Kennedy_Jim@bls.gov ABSTRACT The particle swarm algorithm is usually a dynamic process, where a point in the search space to be tested depends on the previous point and the direction of movement. The process can be decomposed, and probability distributions around a center can be used instead of the usual trajectory approach. A version that is both dynamic and Gaussian looks very promising. ACM Categories & Subject Descriptors I.2.11 Distributed Artificial Intelligence Multiagent systems Keywords: Particle swarms 1. INTRODUCTION Since its introduction in 1995 (Kennedy and Eberhart, 1995; Eberhart and Kennedy, 1995), the particle swarm algorithm has gone through many changes. Though early results were surprisingly good, and though the method had very few moving parts, it turned out to be quite difficult to understand how it worked, in order to improve it. Over the past decade, numerous modifications have been introduced, several of which have turned out to cause genuine improvements in performance, and several of which have helped to understand the dynamics of the swarm and how it is able to solve problems. The canonical particle swarm algorithm is given as: For each population

There are no references for this article.