El yazısı rakam sınıflandırması için gözetimsiz benzerlik tabanlı evrişimler
Yükleniyor...
Tarih
2022
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Institute of Electrical and Electronics Engineers Inc.
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Effective 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.
Açıklama
Anahtar Kelimeler
Convolutional neural networks, Digit classification, Unsupervised learning, Convolution, Backpropagation training, Classification results, CNN filters, Convolutional neural network, Fine tuning, Handwritten digit classification, Hyper-parameter, Initialization methods, Unsupervised method, Object detection, Deep learning, IOU
Kaynak
WoS Q Değeri
Scopus Q Değeri
Cilt
Sayı
Künye
Erkoç, 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/SIU55565.2022.9864689
