Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies

Learning Bayesian Networks (BNs) from high-dimensional data is a complex and time-consuming task. Although there are approaches based on horizontal (instances) or vertical (variables) partitioning in the literature, none can guarantee the same theoretical properties as the Greedy Equivalence Search...

Descripción completa

Detalles Bibliográficos
Autores: Laborda Sicilia, Jorge Daniel, Torrijos Arenas, Pablo, Puerta Callejón, José Miguel, Gámez Martín, José Antonio
Tipo de recurso: artículo
Fecha de publicación:2024
País:España
Institución:Universidad de Castilla-La Mancha
Repositorio:RUIdeRA. Repositorio Institucional de la UCLM
OAI Identifier:oai:ruidera.uclm.es:10578/43411
Acceso en línea:https://doi.org/10.1016/j.ijar.2024.109302
https://www.sciencedirect.com/science/article/pii/S0888613X24001890?via%3Dihub
https://hdl.handle.net/10578/43411
Access Level:acceso abierto
Palabra clave:Bayesian Network fusion/aggregation
Bayesian Network learning
Distributed machine learning
id ES_c3a577800f75b7bcffaafee8081b726a
oai_identifier_str oai:ruidera.uclm.es:10578/43411
network_acronym_str ES
network_name_str España
repository_id_str
spelling Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologiesLaborda Sicilia, Jorge DanielTorrijos Arenas, PabloPuerta Callejón, José MiguelGámez Martín, José AntonioBayesian Network fusion/aggregationBayesian Network learningDistributed machine learningLearning Bayesian Networks (BNs) from high-dimensional data is a complex and time-consuming task. Although there are approaches based on horizontal (instances) or vertical (variables) partitioning in the literature, none can guarantee the same theoretical properties as the Greedy Equivalence Search (GES) algorithm, except those based on the GES algorithm itself. This paper proposes a parallel distributed framework that uses GES as its local learning algorithm, obtaining results similar to those of GES and guaranteeing its theoretical properties but requiring less execution time. The framework involves splitting the set of all possible edges into clusters and constraining each framework node to only work with the received subset of edges. The global learning process is an iterative algorithm that carries out rounds until a convergence criterion is met. We have designed a ring and a star topology to distribute node connections. Regardless of the topology, each node receives a BN as input; it then fuses it with its own BN model and uses the result as the starting point for a local learning process, limited to its own subset of edges. Once finished, the result is then sent to another node as input. Experiments were carried out on a large repertory of domains, including large BNs up to more than 1000 variables. Our results demonstrate our proposals effectiveness compared to GES and its fast version (fGES), generating high-quality BNs in less execution time.Elsevier202520252024info:eu-repo/semantics/articleapplication/pdfapplication/pdfhttps://doi.org/10.1016/j.ijar.2024.109302https://www.sciencedirect.com/science/article/pii/S0888613X24001890?via%3Dihubhttps://hdl.handle.net/10578/43411reponame:RUIdeRA. Repositorio Institucional de la UCLMinstname:Universidad de Castilla-La ManchaEspañolSBPLY/21/180225/000062MCIN/AEI/10.13039/501100011033PID2022-139293NB-C32TED2021-131291B-I00FPU21/010742022-GRIN-344372019-PREDUCLM-10188info:eu-repo/semantics/openAccessoai:ruidera.uclm.es:10578/434112026-05-27T07:36:41Z
dc.title.none.fl_str_mv Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
title Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
spellingShingle Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
Laborda Sicilia, Jorge Daniel
Bayesian Network fusion/aggregation
Bayesian Network learning
Distributed machine learning
title_short Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
title_full Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
title_fullStr Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
title_full_unstemmed Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
title_sort Distributed fusion-based algorithms for learning high-dimensional Bayesian Networks: Testing ring and star topologies
dc.creator.none.fl_str_mv Laborda Sicilia, Jorge Daniel
Torrijos Arenas, Pablo
Puerta Callejón, José Miguel
Gámez Martín, José Antonio
author Laborda Sicilia, Jorge Daniel
author_facet Laborda Sicilia, Jorge Daniel
Torrijos Arenas, Pablo
Puerta Callejón, José Miguel
Gámez Martín, José Antonio
author_role author
author2 Torrijos Arenas, Pablo
Puerta Callejón, José Miguel
Gámez Martín, José Antonio
author2_role author
author
author
dc.subject.none.fl_str_mv Bayesian Network fusion/aggregation
Bayesian Network learning
Distributed machine learning
topic Bayesian Network fusion/aggregation
Bayesian Network learning
Distributed machine learning
description Learning Bayesian Networks (BNs) from high-dimensional data is a complex and time-consuming task. Although there are approaches based on horizontal (instances) or vertical (variables) partitioning in the literature, none can guarantee the same theoretical properties as the Greedy Equivalence Search (GES) algorithm, except those based on the GES algorithm itself. This paper proposes a parallel distributed framework that uses GES as its local learning algorithm, obtaining results similar to those of GES and guaranteeing its theoretical properties but requiring less execution time. The framework involves splitting the set of all possible edges into clusters and constraining each framework node to only work with the received subset of edges. The global learning process is an iterative algorithm that carries out rounds until a convergence criterion is met. We have designed a ring and a star topology to distribute node connections. Regardless of the topology, each node receives a BN as input; it then fuses it with its own BN model and uses the result as the starting point for a local learning process, limited to its own subset of edges. Once finished, the result is then sent to another node as input. Experiments were carried out on a large repertory of domains, including large BNs up to more than 1000 variables. Our results demonstrate our proposals effectiveness compared to GES and its fast version (fGES), generating high-quality BNs in less execution time.
publishDate 2024
dc.date.none.fl_str_mv 2024
2025
2025
dc.type.none.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://doi.org/10.1016/j.ijar.2024.109302
https://www.sciencedirect.com/science/article/pii/S0888613X24001890?via%3Dihub
https://hdl.handle.net/10578/43411
url https://doi.org/10.1016/j.ijar.2024.109302
https://www.sciencedirect.com/science/article/pii/S0888613X24001890?via%3Dihub
https://hdl.handle.net/10578/43411
dc.language.none.fl_str_mv Español
language_invalid_str_mv Español
dc.relation.none.fl_str_mv SBPLY/21/180225/000062
MCIN/AEI/10.13039/501100011033
PID2022-139293NB-C32
TED2021-131291B-I00
FPU21/01074
2022-GRIN-34437
2019-PREDUCLM-10188
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:RUIdeRA. Repositorio Institucional de la UCLM
instname:Universidad de Castilla-La Mancha
instname_str Universidad de Castilla-La Mancha
reponame_str RUIdeRA. Repositorio Institucional de la UCLM
collection RUIdeRA. Repositorio Institucional de la UCLM
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869418813567533056
score 15,812429

Similares en LA Referencia