Accelerated learning in multi-layer neural networks

Author

Negnevitsky, Michael ; Ringrose, Martin

Author_Institution

Sch. of Electr. Eng. & Comput. Sci., Tasmania Univ., Hobart, Tas., Australia

Volume

3

fYear

1999

fDate

1999

Firstpage

1167

Abstract

The most popular training method for multi-layer feedforward networks has traditionally been the error backpropagation algorithm. This algorithm has proved to be slow in its convergence to the error minimum; thus, several methods of accelerating learning using backpropagation have been developed. These include using hyperbolic tangent activation functions, momentum, adaptive learning rates and fuzzy control of the learning parameters. These methods are looked at in this paper

Keywords

backpropagation; convergence; feedforward neural nets; fuzzy control; momentum; multilayer perceptrons; transfer functions; accelerated learning; adaptive learning rates; convergence; error backpropagation algorithm; error minimum; fuzzy control; hyperbolic tangent activation functions; learning parameters; momentum; multilayer feedforward neural networks; Acceleration; Australia; Computer errors; Computer science; Convergence; Feedforward systems; Intelligent networks; Multi-layer neural network; Neural networks; Neurons;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Information Processing, 1999. Proceedings. ICONIP '99. 6th International Conference on

Conference_Location

Perth, WA

Print_ISBN

0-7803-5871-6

Type

conf

DOI

10.1109/ICONIP.1999.844701

Filename

844701