Novel hybrid optimization techniques for enhanced generalization and faster convergence in deep learning models: the nestyogi approach to facial biometrics

Basit Öğe Kaydı
dc.contributor.authorAltaher, Raoof
dc.contributor.authorKoyuncu, Hakan
dc.date.accessioned2024-10-23T05:45:02Z
dc.date.available2024-10-23T05:45:02Z
dc.date.issued2024en_US
dc.departmentEnstitüler, Lisansüstü Eğitim Enstitüsü, Elektrik ve Bilgisayar Mühendisliği Ana Bilim Dalıen_US
dc.description.abstractIn the rapidly evolving field of biometric authentication, deep learning has become a cornerstone technology for face detection and recognition tasks. However, traditional optimizers often struggle with challenges such as overfitting, slow convergence, and limited generalization across diverse datasets. To address these issues, this paper introduces NestYogi, a novel hybrid optimization algorithm that integrates the adaptive learning capabilities of the Yogi optimizer, anticipatory updates of Nesterov momentum, and the generalization power of stochastic weight averaging (SWA). This combination significantly improves both the convergence rate and overall accuracy of deep neural networks, even when trained from scratch. Extensive data augmentation techniques, including noise and blur, were employed to ensure the robustness of the model across diverse conditions. NestYogi was rigorously evaluated on two benchmark datasets Labeled Faces in the Wild (LFW) and YouTube Faces (YTF), demonstrating superior performance with a detection accuracy reaching 98% and a recognition accuracy up to 98.6%, outperforming existing optimization strategies. These results emphasize NestYogi’s potential to overcome critical challenges in face detection and recognition, offering a robust solution for achieving state-of-the-art performance in real-world applications.en_US
dc.identifier.citationAltaher, R., Koyuncu, H. (2024). Novel hybrid optimization techniques for enhanced generalization and faster convergence in deep learning models: the nestyogi approach to facial biometrics. Mathematics, 12(18). 10.3390/math12182919en_US
dc.identifier.issn2227-7390
dc.identifier.issue18en_US
dc.identifier.scopus2-s2.0-85205062599
dc.identifier.scopusqualityQ1
dc.identifier.urihttps://hdl.handle.net/20.500.12939/4942
dc.identifier.volume12en_US
dc.identifier.wosWOS:001322966600001
dc.identifier.wosqualityQ1
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.institutionauthorAltaher, Raoof
dc.institutionauthorKoyuncu, Hakan
dc.language.isoen
dc.publisherMultidisciplinary Digital Publishing Institute (MDPI)en_US
dc.relation.ispartofMathematics
dc.relation.isversionof10.3390/math12182919en_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - İdari Personel ve Öğrencien_US
dc.rightsinfo:eu-repo/semantics/openAccessen_US
dc.subjectBiometric Authenticationen_US
dc.subjectFace Detectionen_US
dc.subjectFace Recognitionen_US
dc.subjectNesterov Momentumen_US
dc.subjectStochastic Weight Averaging (SWA)en_US
dc.subjectTriplet Lossen_US
dc.subjectYogi Algorithmen_US
dc.titleNovel hybrid optimization techniques for enhanced generalization and faster convergence in deep learning models: the nestyogi approach to facial biometrics
dc.typeArticle

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim
Ä°sim:
mathematics-12-02919.pdf
Boyut:
23.72 MB
Biçim:
Adobe Portable Document Format
Açıklama:
Tam Metin / Full Text
Ä°ndir
Lisans paketi
Listeleniyor 1 - 1 / 1
[ X ]
Ä°sim:
license.txt
Boyut:
1.44 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
Ä°ndir

Koleksiyon

Scopus İndeksli Yayınlar Koleksiyonu
Tez Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu