Improvements to the Backpropagation Algorithm

This paper presents some simple techniques to improve the backpropagation algorithm. Since learning in neural networks is an NP-complete problem and since traditional gradient descent methods are rather slow, many alternatives have been tried in order to accelerate convergence. Some of the proposed methods are mutually compatible and a combination of them normally works better than each method alone.

MoreLess

Year of publication:	2012
Authors:	Petrini, Mircea
Published in:	Annals of the University of Petrosani, Economics. - University of Petrosani, Romania. - Vol. 12.2012, 4, p. 185-192
Publisher:	University of Petrosani, Romania
Subject:	Artificial Neural Network (ANN) \| backpropagation \| extended network

More details

Extent:	application/pdf
Type of publication:	Article
Classification:	C - Mathematical and Quantitative Methods
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10011067025