Comparison of Different Surrogate Models for the JADE Algorithm

Konrad Krawczyk, Jarosław Arabas

2023

Abstract

We investigate the performance of various regression-based surrogate models integrated with a ranking procedure in the Adaptive Differential Evolution with an Optional External Archive (JADE) method. We perform regression of the fitness function by the surrogate model to reduce the number of fitness evaluations needed to achieve the optimization progress. The surrogate model training process should be relatively cheap since training is performed many times along with the optimization process. Therefore we investigate surrogate models based on k Nearest Neighbors, Random Forests, and Support Vector Machines. We test the effectiveness of JADE with and without the surrogate models using the CEC2013 benchmark set for single-criterion continuous optimization. Experimental data confirm the benefits of using the surrogate models and indicate the difference in efficiency improvement between the considered models.

Paper Citation

in Harvard Style

Krawczyk K. and Arabas J. (2023). Comparison of Different Surrogate Models for the JADE Algorithm. In Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA; ISBN 978-989-758-674-3, SciTePress, pages 186-194. DOI: 10.5220/0012166100003595

in Bibtex Style

@conference{ecta23,
author={Konrad Krawczyk and Jarosław Arabas},
title={Comparison of Different Surrogate Models for the JADE Algorithm},
booktitle={Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA},
year={2023},
pages={186-194},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012166100003595},
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 - Comparison of Different Surrogate Models for the JADE Algorithm
SN - 978-989-758-674-3
AU - Krawczyk K.
AU - Arabas J.
PY - 2023
SP - 186
EP - 194
DO - 10.5220/0012166100003595
PB - SciTePress