Effect of Randomization of Sampling Strategies on Fitness Values Distribution
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F25%3A10260467" target="_blank" >RIV/61989100:27240/25:10260467 - isvavai.cz</a>
Result on the web
<a href="https://ieeexplore.ieee.org/abstract/document/11015357" target="_blank" >https://ieeexplore.ieee.org/abstract/document/11015357</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/ISCBI64586.2025.11015357" target="_blank" >10.1109/ISCBI64586.2025.11015357</a>
Alternative languages
Result language
angličtina
Original language name
Effect of Randomization of Sampling Strategies on Fitness Values Distribution
Original language description
Exploratory Landscape Analysis characterizes fitness landscapes of various optimization problems to categorize them and reveal relationships with different metaheuristic algorithms. It uses finite sample sets to map hypersurfaces defined by fitness values to feature space. Research indicates that the chosen sampling strategy and additional randomization can influence landscape features. Recently, a feature representing each fitness landscape through a normalized histogram of fitness values has been introduced as a simple method. This paper presents an experimental evaluation of an alternative, lightweight method for generating random samples based on the Cranley-Patterson rotation in the context of the fitness histograms. It conducts a cluster analysis of the resulting feature space and assesses the impact of the randomization on the distribution of fitness values.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10200 - Computer and information sciences
Result continuities
Project
<a href="/en/project/GF22-34873K" target="_blank" >GF22-34873K: Constrained Multiobjective Optimization Based on Problem Landscape Analysis</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2025
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of the International Symposium on Computational and Business Intelligence, ISCBI 2025
ISBN
979-8-3315-3379-3
ISSN
2832-4757
e-ISSN
2832-4749
Number of pages
5
Pages from-to
88-92
Publisher name
IEEE
Place of publication
Piscataway
Event location
Macao
Event date
Feb 14, 2025
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—