Please use this identifier to cite or link to this item:
Title: Quaternionic convolutional neural networks with trainable Bessel activation functions
Author: Vieira, Nelson
Keywords: Artificial neural networks and deep learning
Activation functions
Quaternionic convolutional neural networks
Bessel functions
Parametric activation functions
Issue Date: Sep-2023
Publisher: Springer
Abstract: Quaternionic Convolutional Neural Networks (QCNN) possess the ability to capture both external dependencies between neighboring features and internal latent dependencies within features of an input vector. In this study, we employ QCNN with activation functions based on Bessel-type functions with trainable parameters, for performing classification tasks. Our experimental results demonstrate that this activation function outperforms the traditional ReLU activation function. Throughout our simulations, we explore various network architectures. The use of activation functions with trainable parameters offers several advantages, including enhanced flexibility, adaptability, improved learning, customized model behavior, and automatic feature extraction.
Peer review: yes
DOI: 10.1007/s11785-023-01387-z
ISSN: 1661-8254
Appears in Collections:CIDMA - Artigos
DMat - Artigos
CHAG - Artigos

Files in This Item:
File Description SizeFormat 
Quaternionic convolutional neural networks with trainable Bessel activation functions.pdf1.14 MBAdobe PDFView/Open

Formato BibTex MendeleyEndnote Degois 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.