Bayesian generation of synthetic datasets for machine-learning tasks: a performance study

Vista sencilla de ítem
dc.contributor.author Fosci, Paolo
dc.contributor.authorNieves Acedo, Javier
dc.contributor.authorPsaila, Giuseppe
dc.contributor.authorBoffelli, Jacopo
dc.contributor.authorGarcía Bringas, Pablo
dc.date.accessioned2026-02-20T16:36:13Z
dc.date.available2026-02-20T16:36:13Z
dc.date.issued2026-03-14
dc.date.updated2026-02-20T16:36:13Z
dc.description.abstractPerforming Machine Learning (ML) tasks on large-scale datasets, as well as simply storing them for subsequent analysis or for long-term archival, require large computational power. The described approach builds on the technique known as “Bayesian Generation” to produce synthetic datasets in such a way that the probability distribution in the source dataset is maintained as much as possible in the new synthetic ones, even if they are much smaller than the original (large) dataset. In fact, this study investigates the impact of generating smaller synthetic datasets for training ML models in place of the original dataset, adopting a twofold perspective. Firstly, the impact on the effectiveness of ML models trained on these smaller synthetic datasets is assessed. Secondly, the amount of computational resources required to generate the synthetic datasets, train ML models on them, and perform the testing phase is measured. Specifically, both execution time and main memory usage are taken into account. Finally, this research work shows that the loss in terms of effectiveness remains consistently limited and stable, and it identifies the scenarios and ML techniques for which incorporating the generation of small synthetic datasets into the ML pipeline can be beneficial for practical deployment in environments with constrained computational resources, such as mobile or industrial devices.en
dc.description.sponsorshipThis study was funded by the European Union - NextGenerationEU, within the framework of the GRINS - Growing Resilient, INclusive and Sustainable project (GRINS PE00000018 – CUP F83C22001720001)en
dc.identifier.citationFosci, P., Nieves, J., Psaila, G., Boffelli, J., & Garcia Bringas, P. (2026). Bayesian generation of synthetic datasets for machine-learning tasks: a performance study. Neurocomputing, 670. https://doi.org/10.1016/J.NEUCOM.2025.132508
dc.identifier.doi10.1016/J.NEUCOM.2025.132508
dc.identifier.eissn1872-8286
dc.identifier.issn0925-2312
dc.identifier.urihttps://hdl.handle.net/20.500.14454/5189
dc.language.isoeng
dc.publisherElsevier B.V.
dc.rights© 2025 The Author(s)
dc.subject.otherBayesian generation
dc.subject.otherBayesian networks
dc.subject.otherEffectiveness and efficiency
dc.subject.otherGeneration of synthetic data
dc.subject.otherThe YABaGen tool
dc.titleBayesian generation of synthetic datasets for machine-learning tasks: a performance studyen
dc.typejournal article
dcterms.accessRightsopen access
oaire.citation.titleNeurocomputing
oaire.citation.volume670
oaire.licenseConditionhttps://creativecommons.org/licenses/by/4.0/
oaire.versionVoR
Archivos
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
fosci_bayesian_2026.pdf
Tamaño:
2.97 MB
Formato:
Adobe Portable Document Format
Descargar
Colecciones
Artículos