Title :
A fast training algorithm for neural networks
Author :
Bilski, Jaroslaw ; Rutkowski, Leszek
Author_Institution :
Dept. of Comput. Eng., Tech. Univ. of Czestochowa, Poland
fDate :
6/1/1998 12:00:00 AM
Abstract :
The recursive least squares method (RLS) is derived for the learning of multilayer feedforward neural networks. Simulation results on the XOR, 4-2-4 encoder, and function approximation problems indicate a fast learning process in comparison to the classical and momentum backpropagation (BP) algorithms
Keywords :
encoding; feedforward neural nets; function approximation; learning (artificial intelligence); least squares approximations; 4-2-4 encoder problem; RLS method; XOR problem; fast learning process; fast training algorithm; function approximation problem; multilayer feedforward neural networks; neural networks; recursive least squares method; Backpropagation algorithms; Feedforward neural networks; Function approximation; Least squares methods; Multi-layer neural network; Neural networks; Neurons; Resonance light scattering; Terminology; Vectors;
Journal_Title :
Circuits and Systems II: Analog and Digital Signal Processing, IEEE Transactions on
