Analysis of Bayesian Network Learning Techniques for a Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm: a case study on MNK Landscape

Marcella S.R. Martins; Mohamed El Yafrani; Myriam Delgado; Ricardo Lüders; Roberto Santana; Hugo V. Siqueira; Huseyin G. Akcay; Belaïd Ahiod

doi:10.1007/s10732-021-09469-x

Analysis of Bayesian Network Learning Techniques for a Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm: a case study on MNK Landscape

Marcella S.R. Martins^*
, Mohamed El Yafrani
, Myriam Delgado
, Ricardo Lüders
, Roberto Santana
, Hugo V. Siqueira
, Huseyin G. Akcay
, Belaïd Ahiod

^*Corresponding author for this work

Research output: Contribution to journal › Journal article › Research › peer-review

13 Citations (Scopus)

Abstract

This work investigates different Bayesian network structure learning techniques by thoroughly studying several variants of Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm (HMOBEDA), applied to the MNK Landscape combinatorial problem. In the experiments, we evaluate the performance considering three different aspects: optimization abilities, robustness and learning efficiency. Results for instances of multi- and many-objective MNK-landscape show that, score-based structure learning algorithms appear to be the best choice. In particular, HMOBEDA_k₂ was capable of producing results comparable with the other variants in terms of the runtime of convergence and the coverage of the final Pareto front, with the additional advantage of providing solutions that are less sensible to noise while the variability of the corresponding Bayesian network models is reduced.

Original language	English
Journal	Journal of Heuristics
Volume	27
Issue number	4
Pages (from-to)	549-573
Number of pages	25
ISSN	1381-1231
DOIs	https://doi.org/10.1007/s10732-021-09469-x
Publication status	Published - Aug 2021

Bibliographical note

Funding Information:
M. Delgado acknowledges CNPq, grants 309935/2017-2 e 439226/2018-0. R. Santana acknowledges support by the TIN2016-78365-R (Spanish Ministry of Economy, Industry and Competitiveness), PID2019-104966GB-I00 (Spanish Ministry of Science and Innovation), the IT-1244-19 (Basque Government) program and project 3KIA (KK-2020/00049) funded by the SPRI-Basque Government through the ELKARTEK program.

Publisher Copyright:
© 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC part of Springer Nature.

Keywords

Estimation of distribution algorithms
Many-objective optimization
Robustness
Structure learning techniques

Access to Document

10.1007/s10732-021-09469-x

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@article{1738a16882004a4b82899a3258f6df4b,

title = "Analysis of Bayesian Network Learning Techniques for a Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm: a case study on MNK Landscape",

abstract = "This work investigates different Bayesian network structure learning techniques by thoroughly studying several variants of Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm (HMOBEDA), applied to the MNK Landscape combinatorial problem. In the experiments, we evaluate the performance considering three different aspects: optimization abilities, robustness and learning efficiency. Results for instances of multi- and many-objective MNK-landscape show that, score-based structure learning algorithms appear to be the best choice. In particular, HMOBEDAk2 was capable of producing results comparable with the other variants in terms of the runtime of convergence and the coverage of the final Pareto front, with the additional advantage of providing solutions that are less sensible to noise while the variability of the corresponding Bayesian network models is reduced.",

keywords = "Estimation of distribution algorithms, Many-objective optimization, Robustness, Structure learning techniques",

author = "Martins, \{Marcella S.R.\} and \{El Yafrani\}, Mohamed and Myriam Delgado and Ricardo L{\"u}ders and Roberto Santana and Siqueira, \{Hugo V.\} and Akcay, \{Huseyin G.\} and Bela{\"i}d Ahiod",

note = "Funding Information: M. Delgado acknowledges CNPq, grants 309935/2017-2 e 439226/2018-0. R. Santana acknowledges support by the TIN2016-78365-R (Spanish Ministry of Economy, Industry and Competitiveness), PID2019-104966GB-I00 (Spanish Ministry of Science and Innovation), the IT-1244-19 (Basque Government) program and project 3KIA (KK-2020/00049) funded by the SPRI-Basque Government through the ELKARTEK program. Publisher Copyright: {\textcopyright} 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC part of Springer Nature.",

year = "2021",

month = aug,

doi = "10.1007/s10732-021-09469-x",

language = "English",

volume = "27",

pages = "549--573",

journal = "Journal of Heuristics",

issn = "1381-1231",

publisher = "Springer",

number = "4",

}

Analysis of Bayesian Network Learning Techniques for a Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm: a case study on MNK Landscape. / Martins, Marcella S.R.; El Yafrani, Mohamed; Delgado, Myriam et al.
In: Journal of Heuristics, Vol. 27, No. 4, 08.2021, p. 549-573.

Research output: Contribution to journal › Journal article › Research › peer-review

TY - JOUR

T1 - Analysis of Bayesian Network Learning Techniques for a Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm

T2 - a case study on MNK Landscape

AU - Martins, Marcella S.R.

AU - El Yafrani, Mohamed

AU - Delgado, Myriam

AU - Lüders, Ricardo

AU - Santana, Roberto

AU - Siqueira, Hugo V.

AU - Akcay, Huseyin G.

AU - Ahiod, Belaïd

N1 - Funding Information: M. Delgado acknowledges CNPq, grants 309935/2017-2 e 439226/2018-0. R. Santana acknowledges support by the TIN2016-78365-R (Spanish Ministry of Economy, Industry and Competitiveness), PID2019-104966GB-I00 (Spanish Ministry of Science and Innovation), the IT-1244-19 (Basque Government) program and project 3KIA (KK-2020/00049) funded by the SPRI-Basque Government through the ELKARTEK program. Publisher Copyright: © 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC part of Springer Nature.

PY - 2021/8

Y1 - 2021/8

N2 - This work investigates different Bayesian network structure learning techniques by thoroughly studying several variants of Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm (HMOBEDA), applied to the MNK Landscape combinatorial problem. In the experiments, we evaluate the performance considering three different aspects: optimization abilities, robustness and learning efficiency. Results for instances of multi- and many-objective MNK-landscape show that, score-based structure learning algorithms appear to be the best choice. In particular, HMOBEDAk2 was capable of producing results comparable with the other variants in terms of the runtime of convergence and the coverage of the final Pareto front, with the additional advantage of providing solutions that are less sensible to noise while the variability of the corresponding Bayesian network models is reduced.

AB - This work investigates different Bayesian network structure learning techniques by thoroughly studying several variants of Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm (HMOBEDA), applied to the MNK Landscape combinatorial problem. In the experiments, we evaluate the performance considering three different aspects: optimization abilities, robustness and learning efficiency. Results for instances of multi- and many-objective MNK-landscape show that, score-based structure learning algorithms appear to be the best choice. In particular, HMOBEDAk2 was capable of producing results comparable with the other variants in terms of the runtime of convergence and the coverage of the final Pareto front, with the additional advantage of providing solutions that are less sensible to noise while the variability of the corresponding Bayesian network models is reduced.

KW - Estimation of distribution algorithms

KW - Many-objective optimization

KW - Robustness

KW - Structure learning techniques

UR - https://www.scopus.com/pages/publications/85100676007

U2 - 10.1007/s10732-021-09469-x

DO - 10.1007/s10732-021-09469-x

M3 - Journal article

AN - SCOPUS:85100676007

SN - 1381-1231

VL - 27

SP - 549

EP - 573

JO - Journal of Heuristics

JF - Journal of Heuristics

IS - 4

ER -

Analysis of Bayesian Network Learning Techniques for a Hybrid Multi-objective Bayesian Estimation of Distribution Algorithm: a case study on MNK Landscape

Abstract

Bibliographical note

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this