Detail publikace

Bayesian Optimization Algorithms for Dynamic Problems

KOBLIHA, M. SCHWARZ, J. OČENÁŠEK, J.

Originální název

Bayesian Optimization Algorithms for Dynamic Problems

Typ

článek ve sborníku mimo WoS a Scopus

Jazyk

angličtina

Originální abstrakt

Thispaper is an experimental study investigating the capability ofBayesian optimization algorithms to solve dynamic problems. We testedthe performance of two variants of Bayesian optimization algorithms -Mixed continuous-discrete Bayesian Optimization Algorithm (MBOA),Adaptive Mixed Bayesian Optimization Algorithm (AMBOA) - and newproposed modifications with embedded Sentinels concept andHypervariance. We have compared the performance of these variants ona simple dynamic problem - a time-varying function with predefinedparameters. The experimental resultsconfirmed the benefit of Sentinels concept and Hypervariance embeddedinto MBOA algorithm for tracking a moving optimum.

Klíčová slova

BOA algorithm, dynamic problem, optimalization

Autoři

KOBLIHA, M.; SCHWARZ, J.; OČENÁŠEK, J.

Rok RIV

2006

Vydáno

8. 3. 2006

Nakladatel

Springer Verlag

Místo

Budapest

ISBN

3-540-33237-5

Kniha

Applications of Evolutionary Computing

ISSN

0302-9743

Periodikum

Lecture Notes in Computer Science

Ročník

2006

Číslo

3907

Stát

Spolková republika Německo

Strany od

800

Strany do

804

Strany počet

5

BibTex

@inproceedings{BUT22416,
  author="Miloš {Kobliha} and Josef {Schwarz} and Jiří {Očenášek}",
  title="Bayesian Optimization Algorithms for Dynamic Problems",
  booktitle="Applications of Evolutionary Computing",
  year="2006",
  journal="Lecture Notes in Computer Science",
  volume="2006",
  number="3907",
  pages="800--804",
  publisher="Springer Verlag",
  address="Budapest",
  isbn="3-540-33237-5",
  issn="0302-9743"
}