El yazısı rakam sınıflandırması için gözetimsiz benzerlik tabanlı evrişimler

Basit Öğe Kaydı
dc.authorid0000-0001-9033-8934
dc.authorid0000-0003-0298-0690
dc.contributor.authorErkoç, Tuğbaen_US
dc.contributor.authorEskil, Mustafa Taneren_US
dc.date.accessioned2022-10-31T16:28:43Z
dc.date.available2022-10-31T16:28:43Z
dc.date.issued2022
dc.departmentIşık Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Bilgisayar Mühendisliği Bölümüen_US
dc.departmentIşık University, Faculty of Engineering and Natural Sciences, Department of Computer Engineeringen_US
dc.description.abstractEffective training of filters in Convolutional Neural Networks (CNN) ensures their success. In order to achieve good classification results in CNNs, filters must be carefully initialized, trained and fine-tuned. We propose an unsupervised method that allows the discovery of filters from the given dataset in a single epoch without specifying the number of filters hyper-parameter in convolutional layers. Our proposed method gradually builds the convolutional layers by a discovery routine that extracts a number of features that adequately represent the complexity of the input domain. The discovered filters represent the patterns in the domain, so they do not require any initialization method or backpropagation training for fine tuning purposes. Our method achieves 99.03% accuracy on MNIST dataset without applying any data augmentation techniques.en_US
dc.description.versionPublisher's Versionen_US
dc.identifier.citationErkoç, T. & Eskil, M. T. (2022). El yazısı rakam sınıflandırması için gözetimsiz benzerlik tabanlı evrişimler. Paper presented at the 2022 30th Signal Processing and Communications Applications Conference (SIU), 1-4. doi:10.1109/SIU55565.2022.9864689en_US
dc.identifier.endpage4
dc.identifier.isbn9781665450928
dc.identifier.isbn9781665450935
dc.identifier.issn2165-0608
dc.identifier.startpage1
dc.identifier.urihttps://hdl.handle.net/11729/5101
dc.identifier.urihttp://dx.doi.org/10.1109/SIU55565.2022.9864689
dc.indekslendigikaynakScopusen_US
dc.institutionauthorErkoç, Tuğbaen_US
dc.institutionauthorEskil, Mustafa Taneren_US
dc.institutionauthorid0000-0001-9033-8934
dc.institutionauthorid0000-0003-0298-0690
dc.language.isotren_US
dc.peerreviewedYesen_US
dc.publicationstatusPublisheden_US
dc.publisherInstitute of Electrical and Electronics Engineers Inc.en_US
dc.relation.isversionof10.1109/SIU55565.2022.9864689
dc.relation.journal2022 30th Signal Processing and Communications Applications Conference (SIU)en_US
dc.relation.publicationcategoryKonferans Öğesi - Uluslararası - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectConvolutional neural networksen_US
dc.subjectDigit classificationen_US
dc.subjectUnsupervised learningen_US
dc.subjectConvolutionen_US
dc.subjectBackpropagation trainingen_US
dc.subjectClassification resultsen_US
dc.subjectCNN filtersen_US
dc.subjectConvolutional neural networken_US
dc.subjectFine tuningen_US
dc.subjectHandwritten digit classificationen_US
dc.subjectHyper-parameteren_US
dc.subjectInitialization methodsen_US
dc.subjectUnsupervised methoden_US
dc.subjectObject detectionen_US
dc.subjectDeep learningen_US
dc.subjectIOUen_US
dc.titleEl yazısı rakam sınıflandırması için gözetimsiz benzerlik tabanlı evrişimleren_US
dc.title.alternativeUnsupervised similarity based convolutions for handwritten digit classificationen_US
dc.typeConference Objecten_US

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
El_yazisi_rakam_siniflandirmasi_icin_gozetimsiz_benzerlik_tabanli_evrisimler.pdf
Boyut:
1006.1 KB
Biçim:
Adobe Portable Document Format
Açıklama:
Publisher's Version
İndir
Lisans paketi
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
license.txt
Boyut:
1.44 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

Bildiri Koleksiyonu | Bilgisayar Mühendisliği Bölümü / Department of Computer Engineering
Scopus İndeksli Yayınlar Koleksiyonu