Alternative Step-Size Adaptation Rule for the Matrix Adaptation Evolution Strategy

Eryk Warchulski, Jarosław Arabas

2024

Abstract

In this paper, we present a comparison of various step-size adaptation rules for the Matrix Adaptation Evolution Strategy (MA-ES) algorithm, which is a lightweight version of the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). In contrast to CMA-ES, MA-ES does not require to invoke numerically complex covariance matrix factorization. We take a step further in this direction and provide a quantitative assessment of alternative step-size rules to Cumulative Step Adaptation (CSA), which is considered to be a state-of-the-art method. Our study shows that generalized 1/5-th success rules like the Previous Population Midpoint Fitness rule (PPMF) or Population Success Rule (PSR) exhibit comparable or superior performance to the CSA rule on standard benchmark problems, including the CEC benchmark suites.

Paper Citation

in Harvard Style

Warchulski E. and Arabas J. (2024). Alternative Step-Size Adaptation Rule for the Matrix Adaptation Evolution Strategy. In Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: ECTA; ISBN 978-989-758-721-4, SciTePress, pages 151-158. DOI: 10.5220/0013012800003837

in Bibtex Style

@conference{ecta24,
author={Eryk Warchulski and Jarosław Arabas},
title={Alternative Step-Size Adaptation Rule for the Matrix Adaptation Evolution Strategy},
booktitle={Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: ECTA},
year={2024},
pages={151-158},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013012800003837},
isbn={978-989-758-721-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: ECTA
TI - Alternative Step-Size Adaptation Rule for the Matrix Adaptation Evolution Strategy
SN - 978-989-758-721-4
AU - Warchulski E.
AU - Arabas J.
PY - 2024
SP - 151
EP - 158
DO - 10.5220/0013012800003837
PB - SciTePress