A method for finding the maximally robust weights of a feed forward neural network with step function neurons

Author

Shonkwiler, Ron ; Meddin, Mona ; Bartz, Michael

Author_Institution

Georgia Inst. of Technol., Atlanta, GA, USA

Volume

3

fYear

1992

fDate

7-11 Jun 1992

Firstpage

153

Abstract

The authors consider a method by which backpropagation is used to find the appropriate interpolating parameters in a network with hard limiter transfer functions. This method has two features: first, it defines a family of sigmoidal functions which converge to the hard limiter so that BP may be used as the appropriate training method; and second, it finds the place to start so that gradient descent yields the absolute minimum and does not become trapped in a relative minimum. The interpolating parameters yield a maximally robust solution

Keywords

backpropagation; feedforward neural nets; learning (artificial intelligence); transfer functions; absolute minimum; backpropagation; feedforward neural network; gradient descent; hard limiter transfer functions; interpolating parameters; maximally robust solution; maximally robust weights; sigmoidal functions; step function neurons; Appropriate technology; Backpropagation; Feedforward neural networks; Feeds; Hardware; Interpolation; Neural networks; Neurons; Robustness; Transfer functions;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1992. IJCNN., International Joint Conference on

Conference_Location

Baltimore, MD

Print_ISBN

0-7803-0559-0

Type

conf

DOI

10.1109/IJCNN.1992.227177

Filename

227177