A binary particle swarm optimization-based pruning approach for environmentally sustainable and robust CNNs

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorUniversity of Sfax-
Autor(es): dc.contributorEconomiques et de Gestion de Jendouba-
Autor(es): dc.contributorPolytech-Sfax (IPSAS)-
Autor(es): dc.contributorUniversidade Estadual Paulista (UNESP)-
Autor(es): dc.contributorEdinburgh Napier University-
Autor(es): dc.creatorTmamna, Jihene-
Autor(es): dc.creatorFourati, Rahma-
Autor(es): dc.creatorBen Ayed, Emna-
Autor(es): dc.creatorPassos, Leandro A.-
Autor(es): dc.creatorPapa, João P.-
Autor(es): dc.creatorBen Ayed, Mounir-
Autor(es): dc.creatorHussain, Amir-
Data de aceite: dc.date.accessioned2025-08-21T18:32:03Z-
Data de disponibilização: dc.date.available2025-08-21T18:32:03Z-
Data de envio: dc.date.issued2025-04-29-
Data de envio: dc.date.issued2024-11-30-
Fonte completa do material: dc.identifierhttp://dx.doi.org/10.1016/j.neucom.2024.128378-
Fonte completa do material: dc.identifierhttps://hdl.handle.net/11449/297298-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/11449/297298-
Descrição: dc.descriptionDeep Convolutional Neural Networks (CNNs), continue to demonstrate remarkable performance across various tasks. However, their computational demands and energy consumption present significant drawbacks, restricting their practical deployment and contributing to a substantial carbon footprint. This paper addresses this challenge by proposing a novel method named Binary Particle Swarm Optimization Layer Pruner (BPSO-LPruner), aimed at achieving substantial computational reduction and mitigating environmental impact during CNN inference. BPSO-LPruner utilizes a constrained Binary Particle Swarm Optimization for CNN layer pruning, integrating a masked-bit strategy and a new population initialization strategy to enhance search performance. We illustrate the effectiveness of our method in reducing model computational costs and carbon footprint emissions while improving performance across multiple models (VGG16, VGG19, DenseNet-40, ResNet18, ResNet20, ResNet34, ResNet44, ResNet56, ResNet110, ResNet50, and MobileNetv2) and diverse datasets (CIFAR-10, CIFAR-100, Tiny-ImageNet, COVID-19 X-ray dataset). Promising results underscore the performance of the proposed method. Additionally, we demonstrate that layer pruning yields benefits beyond enhanced computational performance. Our experimentation reveals that BPSO-LPruner enhances the model's reliability and robustness by effectively addressing variations in input data, inherent ambiguity in model parameters, and adversarial images.-
Descrição: dc.descriptionEngineering and Physical Sciences Research Council-
Descrição: dc.descriptionResearch Groups in Intelligent Machines National Engineering School of Sfax (ENIS) University of Sfax-
Descrição: dc.descriptionUniversité de Jendouba Faculté des Sciences Juridiques Economiques et de Gestion de Jendouba-
Descrição: dc.descriptionIndustry 4.0 Research Lab Polytech-Sfax (IPSAS), Avenue 5 August, Rue Said Aboubaker-
Descrição: dc.descriptionSchool of Sciences São Paulo State University-
Descrição: dc.descriptionComputer Sciences and Communication Department Faculty of Sciences of Sfax University of Sfax-
Descrição: dc.descriptionSchool of Computing Edinburgh Napier University-
Descrição: dc.descriptionSchool of Sciences São Paulo State University-
Descrição: dc.descriptionEngineering and Physical Sciences Research Council: EP/M026981/1-
Descrição: dc.descriptionEngineering and Physical Sciences Research Council: EP/T021063/1-
Descrição: dc.descriptionEngineering and Physical Sciences Research Council: EP/T024917/1-
Idioma: dc.languageen-
Relação: dc.relationNeurocomputing-
???dc.source???: dc.sourceScopus-
Palavras-chave: dc.subjectAdversarial attacks-
Palavras-chave: dc.subjectBinary particle swarm optimization-
Palavras-chave: dc.subjectBit mask strategy-
Palavras-chave: dc.subjectGreen deep learning-
Palavras-chave: dc.subjectLayer pruning-
Palavras-chave: dc.subjectLayer weighting initialization-
Título: dc.titleA binary particle swarm optimization-based pruning approach for environmentally sustainable and robust CNNs-
Tipo de arquivo: dc.typelivro digital-
Aparece nas coleções:Repositório Institucional - Unesp

Não existem arquivos associados a este item.