DyPrune: dynamic pruning rates for neural networks

Jonker, Richard A.A.; Poudel, Roshan; Fajarda, Olga; Oliveira, José Luís; Lopes, Rui Pedro; Matos, Sérgio

Publication

DyPrune: dynamic pruning rates for neural networks

2023Conference paper

dc.contributor.author	Jonker, Richard A.A.
dc.contributor.author	Poudel, Roshan
dc.contributor.author	Fajarda, Olga
dc.contributor.author	Oliveira, José Luís
dc.contributor.author	Lopes, Rui Pedro
dc.contributor.author	Matos, Sérgio
dc.date.accessioned	2024-03-11T09:20:33Z
dc.date.available	2024-03-11T09:20:33Z
dc.date.issued	2023
dc.description.abstract	Neural networks have achieved remarkable success in various applications such as image classification, speech recognition, and natural language processing. However, the growing size of neural networks poses significant challenges in terms of memory usage, computational cost, and deployment on resource-constrained devices. Pruning is a popular technique to reduce the complexity of neural networks by removing unnecessary connections, neurons, or filters. In this paper, we present novel pruning algorithms that can reduce the number of parameters in neural networks by up to 98% without sacrificing accuracy. This is done by scaling the pruning rate of the models to the size of the model and scheduling the pruning to execute throughout the training of the model. Code related to this work is openly available.	pt_PT
dc.description.sponsorship	This work was supported by national funds through the Foundation for Science and Technology (FCT) in the context of the project DSAIPA/AI/0088/2020 and project UIDB/00127/2020.	pt_PT
dc.description.version	info:eu-repo/semantics/publishedVersion	pt_PT
dc.identifier.citation	Jonker, Richard A.A.; Poudel, Roshan; Fajarda, Olga; Oliveira, José Luís; Lopes, Rui Pedro; Matos, Sérgio (2023). DyPrune: dynamic pruning rates for neural networks. In Progress in Artificial Intelligence (EPIA). Cham: Springer. 14115. p. 146-157. ISBN 978-3-031-49007-1	pt_PT
dc.identifier.doi	10.1007/978-3-031-49008-8_12	pt_PT
dc.identifier.isbn	978-3-031-49007-1
dc.identifier.uri	http://hdl.handle.net/10198/29602
dc.language.iso	eng	pt_PT
dc.peerreviewed	yes	pt_PT
dc.publisher	Springer Nature	pt_PT
dc.relation	Institute of Electronics and Informatics Engineering of Aveiro
dc.rights.uri	http://creativecommons.org/licenses/by/4.0/	pt_PT
dc.subject	Machine learning	pt_PT
dc.subject	Neural networks	pt_PT
dc.subject	Pruning	pt_PT
dc.title	DyPrune: dynamic pruning rates for neural networks	pt_PT
dc.type	conference paper
dspace.entity.type	Publication
oaire.awardTitle	Institute of Electronics and Informatics Engineering of Aveiro
oaire.awardURI	info:eu-repo/grantAgreement/FCT/FCT_DSAIPA_2020/DSAIPA%2FAI%2F0088%2F2020/PT
oaire.awardURI	info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F00127%2F2020/PT
oaire.citation.endPage	157	pt_PT
oaire.citation.startPage	146	pt_PT
oaire.citation.title	Progress in Artificial Intelligence (EPIA)	pt_PT
oaire.citation.volume	14115	pt_PT
oaire.fundingStream	FCT_DSAIPA_2020
oaire.fundingStream	6817 - DCRRNI ID
person.familyName	Lopes
person.givenName	Rui Pedro
person.identifier.ciencia-id	8E14-54E4-4DB5
person.identifier.orcid	0000-0002-9170-5078
project.funder.identifier	http://doi.org/10.13039/501100001871
project.funder.identifier	http://doi.org/10.13039/501100001871
project.funder.name	Fundação para a Ciência e a Tecnologia
project.funder.name	Fundação para a Ciência e a Tecnologia
rcaap.rights	restrictedAccess	pt_PT
rcaap.type	conferenceObject	pt_PT
relation.isAuthorOfPublication	e1e64423-0ec8-46ee-be96-33205c7c98a9
relation.isAuthorOfPublication.latestForDiscovery	e1e64423-0ec8-46ee-be96-33205c7c98a9
relation.isProjectOfPublication	b95c01b9-09de-4866-bb86-ac938ce3e0d3
relation.isProjectOfPublication	af5198a3-6377-4245-a289-6656629b8bfa
relation.isProjectOfPublication.latestForDiscovery	b95c01b9-09de-4866-bb86-ac938ce3e0d3