Implementation of neural network with approximations functions

Autor:	G. Lamarque, M. Hnatiuc
Rok vydání:	2004
Předmět:	Programmable logic device Approximation theory symbols.namesake Artificial neural network Gaussian function symbols Multiplication Sigmoid function Algorithm Microsoft Visual Studio Exponential function Mathematics
Zdroj:	SCS 2003. International Symposium on Signals, Circuits and Systems. Proceedings (Cat. No.03EX720).
DOI:	10.1109/scs.2003.1227112
Popis:	The purpose of this work is to stimulate a neural network with non-linear activation functions. The non-linear functions are simulated in Microsoft Visual Studio C++ 6.0 to observe the precision and to implement on the programmable logic devices. This network is realized to accept very small input values. The multiplication between input values and weight values is realized with the add-logarithm and exponential functions. One approximates all the non-linear functions with linear functions using shift-add blocks.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::d9b5e74716148c9c86715773c91f0301 https://doi.org/10.1109/scs.2003.1227112 Zobrazit plný text záznamu