An evolutionary swarm intelligence optimizer based on probabilistic distribution

In this study, we propose a novel approach to balance exploitation and exploration. The proposed approach is the Evolutionary Swarm Intelligence (ESI) optimizer, which combines an exploration-biased strategy with an exploitation-biased operator. The algorithm is built based on the collective behavio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Neural computing & applications Jg. 37; H. 27; S. 22387 - 22399
Hauptverfasser: Yang, Yifei, Yang, Haichuan, Li, Haotian, Tang, Zheng, Gao, Shangce
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London Springer London 01.09.2025
Springer Nature B.V
Schlagworte:
Algorithms
Artificial Intelligence
Biological activity
Biological evolution
Computational Biology/Bioinformatics
Computational Science and Engineering
Computer Science
Data Mining and Knowledge Discovery
Evolution
Exploitation
Genes
Genetic algorithms
Image Processing and Computer Vision
Manual workers
Optimization
Performance evaluation
Probability and Statistics in Computer Science
S.I.: Hybrid Approaches to Nature-inspired Optimization Algorithms and Their Applications
Sociology
Special Issue on Hybrid Approaches to Nature-inspired Optimization Algorithms and Their Applications
Swarm intelligence
Swarm intelligence
Dendritic neuron model
Meta-heuristic algorithms
Genetic algorithm
Exploitation and exploration
ISSN:0941-0643, 1433-3058
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this study, we propose a novel approach to balance exploitation and exploration. The proposed approach is the Evolutionary Swarm Intelligence (ESI) optimizer, which combines an exploration-biased strategy with an exploitation-biased operator. The algorithm is built based on the collective behavior of biological groups, imitating their intelligence behavior. The biological evolutionary process, inspired by genetic algorithms, is applied to every individual in the algorithm. Both swarm intelligence and genetic algorithms have been widely used in practical problems, and their reliability has been proven. ESI is characterized by both spatial group intelligence behavior and temporal biological evolution. To test the performance of ESI, we used a classic test set from IEEE CEC2017 and 22 practical problems from IEEE CEC2011. The popular training tests of the dendritic neuron model were also included in the control trials. We compared ESI with some typical swarm intelligence algorithms and classic algorithms to evaluate its performance and ability to solve practical problems. The experimental results show that ESI outperforms other algorithms in terms of basic performance and the ability to solve practical problems.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0941-0643
1433-3058
DOI:10.1007/s00521-023-09299-x