Please use this identifier to cite or link to this item: http://hdl.handle.net/20.500.12258/18567
Title: Optimization of neural network training for image recognition based on trigonometric polynomial approximation
Authors: Vershkov, N. A.
Вершков, Н. А.
Babenko, M. G.
Бабенко, М. Г.
Kuchukov, V. A.
Кучуков, В. А.
Kuchukova, N. N.
Кучукова, Н. Н.
Keywords: Image recognition;Polynomial approximation;Neural networks
Issue Date: 2021
Publisher: Pleiades journals
Citation: Vershkov, N., Babenko, M., Tchernykh, A., Pulido-Gaytan B., Cortés-Mendoza J.M., Kuchukov, V., Kuchukova, N. Optimization of neural network training for image recognition based on trigonometric polynomial approximation // Programming and Computer Software. - 2021. - Том 47. - Выпуск 8. - Стр.: 830 - 838. - DOI10.1134/S0361768821080272
Series/Report no.: Programming and Computer Software
Abstract: The paper discusses optimization issues of training Artificial Neural Networks (ANNs) using a nonlinear trigonometric polynomial function. The proposed method presents the mathematical model of an ANN as an information transmission system where effective techniques to restore signals are widely used. To optimize ANN training, we use energy characteristics assuming ANNs as data transmission systems. We propose a nonlinear layer in the form of a trigonometric polynomial that approximates the “syncular” function based on the generalized approximation theorem and the wave model. To confirm the theoretical results, the efficiency of the proposed approach is compared with standard ANN implementations with sigmoid and Rectified Linear Unit (ReLU) activation functions. The experimental evaluation shows the same accuracy of standard ANNs with a time reduction of the training phase of supervised learning for the proposed model.
URI: http://hdl.handle.net/20.500.12258/18567
Appears in Collections:Статьи, проиндексированные в SCOPUS, WOS

Files in This Item:
File Description SizeFormat 
scopusresults 1999 .pdf
  Restricted Access
564.25 kBAdobe PDFView/Open
WoS 1358 .pdf
  Restricted Access
85.84 kBAdobe PDFView/Open


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