Show simple item record

dc.contributor.authorVargas Rojas, Víctor Manuel
dc.contributor.authorGutiérrez, Pedro Antonio
dc.contributor.authorHervás-Martínez, César
dc.date.accessioned2021-11-10T12:47:57Z
dc.date.available2021-11-10T12:47:57Z
dc.date.issued2022
dc.identifier.urihttp://hdl.handle.net/10396/22071
dc.description.abstractCurrently, the use of deep learning for solving ordinal classification problems, where categories follow a natural order, has not received much attention. In this paper, we propose an unimodal regularisation based on the beta distribution applied to the cross-entropy loss. This regularisation encourages the distribution of the labels to be a soft unimodal distribution, more appropriate for ordinal problems. Given that the beta distribution has two parameters that must be adjusted, a method to automatically determine them is proposed. The regularised loss function is used to train a deep neural network model with an ordinal scheme in the output layer. The results obtained are statistically analysed and show that the combination of these methods increases the performance in ordinal problems. Moreover, the proposed beta distribution performs better than other distributions proposed in previous works, achieving also a reduced computational cost.es_ES
dc.format.mimetypeapplication/pdfes_ES
dc.language.isoenges_ES
dc.publisherElsevieres_ES
dc.rightshttps://creativecommons.org/licenses/by/4.0/es_ES
dc.sourcePattern Recognition 122, 108310 (2022)es_ES
dc.subjectOrdinal regressiones_ES
dc.subjectUnimodal distributiones_ES
dc.subjectConvolutional networkes_ES
dc.subjectBeta distributiones_ES
dc.subjectStick-breakinges_ES
dc.titleUnimodal regularisation based on beta distribution for deep ordinal regressiones_ES
dc.typeinfo:eu-repo/semantics/articlees_ES
dc.relation.publisherversionhttps://doi.org/10.1016/j.patcog.2021.108310es_ES
dc.relation.projectIDGobierno de España. PID2020-115454GB-C22es_ES
dc.relation.projectIDJunta de Andalucía. UCO-1261651es_ES
dc.relation.projectIDJunta de Andalucía. PY20_00074es_ES
dc.relation.projectIDJunta de Andalucía. PS-2020-780es_ES
dc.relation.projectIDGobierno de España. FPU18/00358es_ES
dc.rights.accessRightsinfo:eu-repo/semantics/openAccesses_ES


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record