Directed random search for multiple layer perceptron training

Author

Seiffert, Udo ; Michaelis, Bernd

Author_Institution

Inst. for Electron. Signal Process. & Commun., Univ. of Magdeburg, Germany

fYear

2001

fDate

2001

Firstpage

193

Lastpage

202

Abstract

Although backpropagation (BP) is commonly used to train multiple layer perceptron (MLP) neural networks and its original algorithm has been significantly improved several times, it still suffers from some drawbacks like being slow, getting stuck in local minima or being bound to constraints regarding the activation (transfer) function of the neurons. The paper presents the substitution of backpropagation with a random search technique which has been enhanced by a directed component. By means of some benchmark problems, a case study shows general potential application fields as well as advantages and disadvantages of both the backpropagation and the directed random search (DRS)

Keywords

backpropagation; multilayer perceptrons; search problems; DRS; MLP neural networks; activation function; backpropagation; case study; directed component; directed random search; local minima; multiple layer perceptron neural networks; multiple layer perceptron training; potential application fields; random search technique; Artificial neural networks; Backpropagation algorithms; Biological system modeling; Biology computing; Computational modeling; Error correction; Neural networks; Neurons; Root mean square; Signal processing algorithms;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing XI, 2001. Proceedings of the 2001 IEEE Signal Processing Society Workshop

Conference_Location

North Falmouth, MA

ISSN

1089-3555

Print_ISBN

0-7803-7196-8

Type

conf

DOI

10.1109/NNSP.2001.943124

Filename

943124