14 June 2000 Conditions of convergence of back-propagation learning algorithm
Author Affiliations +
Proceedings Volume 4148, Optoelectronic and Hybrid Optical/Digital Systems for Image and Signal Processing; (2000) https://doi.org/10.1117/12.388435
Event: International Workshop on Optoelectronic and Hybrid Optical/Digital Systems for Image/Signal Processing, 1999, Lviv, Ukraine
Abstract
In this work properties of various parameters of sigma- function mathematically are described, because of which the influence of these parameters on the speed of convergence of the back-propagation learning algorithm for feedforward neural networks is shown. Because of it, installing optimum conditions and entering some restrictions on parameters the new modified algorithm is received which converges to the necessary solution much faster.
© (2000) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Akylbek Amatovich Jeenbekov, A. A. Sarybaeva, "Conditions of convergence of back-propagation learning algorithm", Proc. SPIE 4148, Optoelectronic and Hybrid Optical/Digital Systems for Image and Signal Processing, (14 June 2000); doi: 10.1117/12.388435; https://doi.org/10.1117/12.388435
PROCEEDINGS
7 PAGES


SHARE
Back to Top