Model complexity vs. performance in the Bayesian optimization algorithm

Elon S. Correa; Jonathan L. Shapiro

doi:10.1007/11844297_101

Model complexity vs. performance in the Bayesian optimization algorithm

Research output: Chapter in Book/Conference proceeding › Conference contribution

Abstract

The Bayesian Optimization Algorithm (BOA) uses a Bayesian network to estimate the probability distribution of promising solutions to a given optimization problem. This distribution is then used to generate new candidate solutions. The objective is to improve the population of candidate solutions by learning and sampling from good solutions. A Bayesian network (BN) is a graphical representation of a probability distribution over a set of variables of a given problem domain. The number of topological states that a BN can create depends on a parameter called maximum allowed indegree. We show that the value of the maximum allowed indegree given to the Bayesian network used by the BOA strongly affects the performance of this algorithm. Furthermore, there is a limited set of values for this parameter for which the performance of the BOA is maximized. © Springer-Verlag Berlin Heidelberg 2006.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)\|Lect. Notes Comput. Sci.
Publisher	Springer Nature
Pages	998-1007
Number of pages	9
Volume	4193
ISBN (Print)	3540389903, 9783540389903
DOIs	https://doi.org/10.1007/11844297_101
Publication status	Published - 2006
Event	9th International Conference on Parallel Problem Solving from Nature, PPSN IX - Reykjavik Duration: 1 Jul 2006 → … http://dblp.uni-trier.de/db/conf/ppsn/ppsn2006.html#CorreaS06http://dblp.uni-trier.de/rec/bibtex/conf/ppsn/CorreaS06.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/ppsn/CorreaS06

Publication series

Name	Lecture Notes in Computer Science

Conference

Conference	9th International Conference on Parallel Problem Solving from Nature, PPSN IX
City	Reykjavik
Period	1/07/06 → …
Internet address	http://dblp.uni-trier.de/db/conf/ppsn/ppsn2006.html#CorreaS06http://dblp.uni-trier.de/rec/bibtex/conf/ppsn/CorreaS06.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/ppsn/CorreaS06

Access to Document

10.1007/11844297_101

Cite this

Correa, E. S., & Shapiro, J. L. (2006). Model complexity vs. performance in the Bayesian optimization algorithm. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci. (Vol. 4193, pp. 998-1007). (Lecture Notes in Computer Science). Springer Nature. https://doi.org/10.1007/11844297_101

@inproceedings{e044c7d41be1447485a81fe7e98371d8,

title = "Model complexity vs. performance in the Bayesian optimization algorithm",

abstract = "The Bayesian Optimization Algorithm (BOA) uses a Bayesian network to estimate the probability distribution of promising solutions to a given optimization problem. This distribution is then used to generate new candidate solutions. The objective is to improve the population of candidate solutions by learning and sampling from good solutions. A Bayesian network (BN) is a graphical representation of a probability distribution over a set of variables of a given problem domain. The number of topological states that a BN can create depends on a parameter called maximum allowed indegree. We show that the value of the maximum allowed indegree given to the Bayesian network used by the BOA strongly affects the performance of this algorithm. Furthermore, there is a limited set of values for this parameter for which the performance of the BOA is maximized. {\textcopyright} Springer-Verlag Berlin Heidelberg 2006.",

author = "Correa, {Elon S.} and Shapiro, {Jonathan L.}",

year = "2006",

doi = "10.1007/11844297_101",

language = "English",

isbn = "3540389903",

volume = "4193",

series = "Lecture Notes in Computer Science",

publisher = "Springer Nature",

pages = "998--1007",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.",

address = "United States",

note = "9th International Conference on Parallel Problem Solving from Nature, PPSN IX ; Conference date: 01-07-2006",

url = "http://dblp.uni-trier.de/db/conf/ppsn/ppsn2006.html#CorreaS06http://dblp.uni-trier.de/rec/bibtex/conf/ppsn/CorreaS06.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/ppsn/CorreaS06",

}

Correa, ES & Shapiro, JL 2006, Model complexity vs. performance in the Bayesian optimization algorithm. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.. vol. 4193, Lecture Notes in Computer Science, Springer Nature, pp. 998-1007, 9th International Conference on Parallel Problem Solving from Nature, PPSN IX, Reykjavik, 1/07/06. https://doi.org/10.1007/11844297_101

Model complexity vs. performance in the Bayesian optimization algorithm. / Correa, Elon S.; Shapiro, Jonathan L.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.. Vol. 4193 Springer Nature, 2006. p. 998-1007 (Lecture Notes in Computer Science).

Research output: Chapter in Book/Conference proceeding › Conference contribution

TY - GEN

T1 - Model complexity vs. performance in the Bayesian optimization algorithm

AU - Correa, Elon S.

AU - Shapiro, Jonathan L.

PY - 2006

Y1 - 2006

N2 - The Bayesian Optimization Algorithm (BOA) uses a Bayesian network to estimate the probability distribution of promising solutions to a given optimization problem. This distribution is then used to generate new candidate solutions. The objective is to improve the population of candidate solutions by learning and sampling from good solutions. A Bayesian network (BN) is a graphical representation of a probability distribution over a set of variables of a given problem domain. The number of topological states that a BN can create depends on a parameter called maximum allowed indegree. We show that the value of the maximum allowed indegree given to the Bayesian network used by the BOA strongly affects the performance of this algorithm. Furthermore, there is a limited set of values for this parameter for which the performance of the BOA is maximized. © Springer-Verlag Berlin Heidelberg 2006.

AB - The Bayesian Optimization Algorithm (BOA) uses a Bayesian network to estimate the probability distribution of promising solutions to a given optimization problem. This distribution is then used to generate new candidate solutions. The objective is to improve the population of candidate solutions by learning and sampling from good solutions. A Bayesian network (BN) is a graphical representation of a probability distribution over a set of variables of a given problem domain. The number of topological states that a BN can create depends on a parameter called maximum allowed indegree. We show that the value of the maximum allowed indegree given to the Bayesian network used by the BOA strongly affects the performance of this algorithm. Furthermore, there is a limited set of values for this parameter for which the performance of the BOA is maximized. © Springer-Verlag Berlin Heidelberg 2006.

U2 - 10.1007/11844297_101

DO - 10.1007/11844297_101

M3 - Conference contribution

SN - 3540389903

SN - 9783540389903

VL - 4193

T3 - Lecture Notes in Computer Science

SP - 998

EP - 1007

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.

PB - Springer Nature

T2 - 9th International Conference on Parallel Problem Solving from Nature, PPSN IX

Y2 - 1 July 2006

ER -

Model complexity vs. performance in the Bayesian optimization algorithm

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this