Paper Info
Title
Parameters' fine tuning of differential evolution algorithm.
Abstract
Most heuristic search method's performances are dependent on parameter choices. These parameter settings govern how new candidate solutions are generated and then applied by the algorithm. They essentially play a key role in determining the quality of the solution obtained and the efficiency of the search. Their fine-tuning techniques are still an on-going research area. Differential Evolution (DE) algorithm is a very powerful optimization method and has become popular in many fields. Based on the prolonged research work on DE, it is now arguably one of the most outstanding stochastic optimization algorithms for real-parameter optimization. One reason for its popularity is its widely appreciated property of having only a small number of parameters to tune. This paper presents a detailed review of DE parameter tuning with a table compromised a recommended guidelines for these parameters, along with a full description of the basic DE algorithm and its corresponding operators, overlooked by previous studies. It is aimed at practitioners to help them achieve better results when adopting DE as an optimization method for their problems with less time and effort. Moreover, an experimental study has been conducted over fifteen test problems and the results obtained prove the reliability of the setting values.
Year
Venue
Keywords
2015
COMPUTER SYSTEMS SCIENCE AND ENGINEERING
Optimization Algorithms,Natural Computing,Evolutionary Algorithms (EAs),Differential Evolution (DE),Parameter Setting,Parameter Tuning,Differential Evolution Mutation,review
Field
DocType
Volume
Computer science,Fine-tuning,Computational science,Differential evolution algorithm,Distributed computing
Journal
30
Issue
ISSN
Citations 
2
0267-6192
1
PageRank 
References 
Authors
0.35
0
3
Name
Order
Citations
PageRank
Rawaa Dawoud Al-Dabbagh1242.13
Saad Mekhilef215925.73
Mohd Sapiyan Baba3517.61