HJADE: An Efficient Differential Evolution with Complex Nonlinear Population Size

Qianyu Zhu, Yifei Yang, Haotian Li, Shibo Dong, Yuki Todo, Shangce Gao

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The differential evolution with optional external archive (JADE) algorithm is a competitive improvement based on differential evolution (DE) algorithm, but it tends to converge to local optima. We propose a new enhancement based on JADE, namely HJADE, which incorporates a strategy of regulating the population size using hybrid functions to strike a more favorable balance between exploration and exploitation. We used 30 problems from the IEEE CEC 2017 benchmark set to assess its implementation and compare HJADE with other JADE-based algorithms using statistical tests. The results show that HJADE exhibits superior performance.

Original languageEnglish
Title of host publicationProceedings - 2023 15th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages67-71
Number of pages5
ISBN (Electronic)9798350326178
DOIs
StatePublished - 2023
Event15th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2023 - Hangzhou, China
Duration: 2023/08/262023/08/27

Publication series

NameProceedings - 2023 15th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2023

Conference

Conference15th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2023
Country/TerritoryChina
CityHangzhou
Period2023/08/262023/08/27

Keywords

  • Differential Evolution
  • Meta-heuristic
  • Optimization
  • Population size

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Human-Computer Interaction
  • Control and Optimization

Fingerprint

Dive into the research topics of 'HJADE: An Efficient Differential Evolution with Complex Nonlinear Population Size'. Together they form a unique fingerprint.

Cite this