The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic Algorithm

Self-adjustment of parameters can significantly improve the performance of evolutionary algorithms. A notable example is the (1 + (λ,λ)) genetic algorithm, where adaptation of the population size helps to achieve the linear running time on the OneMax problem. However, on problems which interfere wit...

Full description

Saved in:
Bibliographic Details
Main Authors: Anton Olegovich Bassin, Maxim Viktorovich Buzdalov, Anatoly Abramovich Shalyto
Format: Article
Language:English
Published: Yaroslavl State University 2020-12-01
Series:Моделирование и анализ информационных систем
Subjects:
Online Access:https://www.mais-journal.ru/jour/article/view/1438
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850025678341144576
author Anton Olegovich Bassin
Maxim Viktorovich Buzdalov
Anatoly Abramovich Shalyto
author_facet Anton Olegovich Bassin
Maxim Viktorovich Buzdalov
Anatoly Abramovich Shalyto
author_sort Anton Olegovich Bassin
collection DOAJ
description Self-adjustment of parameters can significantly improve the performance of evolutionary algorithms. A notable example is the (1 + (λ,λ)) genetic algorithm, where adaptation of the population size helps to achieve the linear running time on the OneMax problem. However, on problems which interfere with the assumptions behind the self-adjustment procedure, its usage can lead to the performance degradation. In particular, this is the case with the “one-fifth rule” on problems with weak fitness-distance correlation.We propose a modification of the “one-fifth rule” in order to have less negative impact on the performance in the cases where the original rule is destructive. Our modification, while still yielding a provable linear runtime on OneMax, shows better results on linear function with random weights, as well as on random satisfiable MAX-3SAT problems.
format Article
id doaj-art-cac3398a2d4741e8a69dfb16134c0df9
institution DOAJ
issn 1818-1015
2313-5417
language English
publishDate 2020-12-01
publisher Yaroslavl State University
record_format Article
series Моделирование и анализ информационных систем
spelling doaj-art-cac3398a2d4741e8a69dfb16134c0df92025-08-20T03:00:45ZengYaroslavl State UniversityМоделирование и анализ информационных систем1818-10152313-54172020-12-0127448850810.18255/1818-1015-2020-4-488-5081094The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic AlgorithmAnton Olegovich Bassin0Maxim Viktorovich Buzdalov1Anatoly Abramovich Shalyto2ITMO UniversityITMO UniversityITMO UniversitySelf-adjustment of parameters can significantly improve the performance of evolutionary algorithms. A notable example is the (1 + (λ,λ)) genetic algorithm, where adaptation of the population size helps to achieve the linear running time on the OneMax problem. However, on problems which interfere with the assumptions behind the self-adjustment procedure, its usage can lead to the performance degradation. In particular, this is the case with the “one-fifth rule” on problems with weak fitness-distance correlation.We propose a modification of the “one-fifth rule” in order to have less negative impact on the performance in the cases where the original rule is destructive. Our modification, while still yielding a provable linear runtime on OneMax, shows better results on linear function with random weights, as well as on random satisfiable MAX-3SAT problems.https://www.mais-journal.ru/jour/article/view/1438parameter adaptation(1 + (λ,λ)) galinear functionsmax-3sat
spellingShingle Anton Olegovich Bassin
Maxim Viktorovich Buzdalov
Anatoly Abramovich Shalyto
The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic Algorithm
Моделирование и анализ информационных систем
parameter adaptation
(1 + (λ,λ)) ga
linear functions
max-3sat
title The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic Algorithm
title_full The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic Algorithm
title_fullStr The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic Algorithm
title_full_unstemmed The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic Algorithm
title_short The “One-fifth Rule” with Rollbacks for Self-Adjustment of the Population Size in the (1 + (λ,λ)) Genetic Algorithm
title_sort one fifth rule with rollbacks for self adjustment of the population size in the 1 λ λ genetic algorithm
topic parameter adaptation
(1 + (λ,λ)) ga
linear functions
max-3sat
url https://www.mais-journal.ru/jour/article/view/1438
work_keys_str_mv AT antonolegovichbassin theonefifthrulewithrollbacksforselfadjustmentofthepopulationsizeinthe1llgeneticalgorithm
AT maximviktorovichbuzdalov theonefifthrulewithrollbacksforselfadjustmentofthepopulationsizeinthe1llgeneticalgorithm
AT anatolyabramovichshalyto theonefifthrulewithrollbacksforselfadjustmentofthepopulationsizeinthe1llgeneticalgorithm
AT antonolegovichbassin onefifthrulewithrollbacksforselfadjustmentofthepopulationsizeinthe1llgeneticalgorithm
AT maximviktorovichbuzdalov onefifthrulewithrollbacksforselfadjustmentofthepopulationsizeinthe1llgeneticalgorithm
AT anatolyabramovichshalyto onefifthrulewithrollbacksforselfadjustmentofthepopulationsizeinthe1llgeneticalgorithm