Adaptive convolution kernel for artificial neural networks

F. Boray Tek*, İlker Çam, Deniz Karlı

*Bu çalışma için yazışmadan sorumlu yazar

Araştırma sonucu: ???type-name???Makalebilirkişi

11 Atıf (Scopus)

Özet

Many deep neural networks are built by using stacked convolutional layers of fixed and single size (often 3 × 3) kernels. This paper describes a method for learning the size of convolutional kernels to provide varying size kernels in a single layer. The method utilizes a differentiable, and therefore backpropagation-trainable Gaussian envelope which can grow or shrink in a base grid. Our experiments compared the proposed adaptive layers to ordinary convolution layers in a simple two-layer network, a deeper residual network, and a U-Net architecture. The results in the popular image classification datasets such as MNIST, MNIST-CLUTTERED, CIFAR-10, Fashion, and “Faces in the Wild” showed that the adaptive kernels can provide statistically significant improvements on ordinary convolution kernels. A segmentation experiment in the Oxford-Pets dataset demonstrated that replacing ordinary convolution layers in a U-shaped network with 7 × 7 adaptive layers can improve its learning performance and ability to generalize.

Orijinal dilİngilizce
Makale numarası103015
DergiJournal of Visual Communication and Image Representation
Hacim75
DOI'lar
Yayın durumuYayınlandı - Şub 2021
Harici olarak yayınlandıEvet

Bibliyografik not

Publisher Copyright:
© 2021 Elsevier Inc.

Parmak izi

Adaptive convolution kernel for artificial neural networks' araştırma başlıklarına git. Birlikte benzersiz bir parmak izi oluştururlar.

Alıntı Yap