A Comparison of the State-of-the-Art Evolutionary Algorithms with Different Stopping Conditions

Jana Herzog, Janez Brest, Borko Bošković

2023

Abstract

This paper focuses on the comparison of the state-of-the-art algorithms and the influence of a stopping condition, the maximum number of function evaluations, on the optimization process. The main aim is to compare the chosen state-of-the-art algorithms with different predetermined stopping conditions and observe if they are comparable in reaching a certain quality of solution on a given set of benchmark functions. For this analysis, the four most recent state-of-the-art evolutionary algorithms were chosen for comparison on the latest set of benchmark functions. We utilized a fixed-budget approach with different values of stopping conditions. Small differences in the algorithms’ performances are observed and the obtained results are also statistically analyzed. Different values of the stopping conditions show different rankings of evolutionary algorithms without the significant difference. The possible reason for this is that their performances are very close.

Paper Citation

in Harvard Style

Herzog J., Brest J. and Bošković B. (2023). A Comparison of the State-of-the-Art Evolutionary Algorithms with Different Stopping Conditions. In Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA; ISBN 978-989-758-674-3, SciTePress, pages 222-229. DOI: 10.5220/0012182200003595

in Bibtex Style

@conference{ecta23,
author={Jana Herzog and Janez Brest and Borko Bošković},
title={A Comparison of the State-of-the-Art Evolutionary Algorithms with Different Stopping Conditions},
booktitle={Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA},
year={2023},
pages={222-229},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012182200003595},
isbn={978-989-758-674-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA
TI - A Comparison of the State-of-the-Art Evolutionary Algorithms with Different Stopping Conditions
SN - 978-989-758-674-3
AU - Herzog J.
AU - Brest J.
AU - Bošković B.
PY - 2023
SP - 222
EP - 229
DO - 10.5220/0012182200003595
PB - SciTePress