.png){kind=icon}
Return
Differential Evolution with Grid-Based Parameter Adaptation
.png){kind=icon}
PRE
AI
DOI：10.1007/s00500-015-1911-2.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The reduction of human intervention in tuning metaheuristic optimization algorithms has been an ongoing research pursuit. Differential Evolution is a very popular algorithm that counts a large number of variants. However, its efficiency has been shown to depend on the type of its crossover operators (binomial or exponential), mutation operators, as well as on the two parameters that dominate these procedures. Making proper decisions on these parameters has proved to be a laborious, problem-dependent task. We propose a parameter adaptation technique that allows the algorithm to dynamically determine the most suitable crossover type and parameter values during its execution. The technique is based on a search procedure in the discretized parameter search space, using estimations of the algorithm's performance. The proposed approach is tested and statistically validated on an established high-dimensional test suite. Also, comparisons with other algorithms are reported, verifying the competitiveness of the proposed approach.
Keywords:
Metaheuristic Optimization
Dynamic Parameter Adaptation
Parameter Control
Differential Evolution
Journal
IF:
2.5
Papers:
1.0W
Citations:
2.1W
Organization
Cited Papers
Editorial scalability of evolutionary algorithms and other metaheuristics for large-scale continuous optimization problems.png)
SOFT COMPUTING
SOFT COMPUTINGIF2.5
PREAI
ShareSaveShareSaveScalability of generalized adaptive differential evolution for large-scale continuous optimizationSOFT COMPUTING
SOFT COMPUTINGIF2.5
PREAI
ShareSaveShareSave
ShareSaveA MOS-based dynamic memetic differential evolution algorithm for continuous optimization: a scalability testSOFT COMPUTING
SOFT COMPUTINGIF2.5
PREAI
ShareSave
View more