Alternative neural network training methods [active sonar processing]

Author

Porto, V.W. ; Fogel, David

Author_Institution

Orincon Corp., San Diego, CA, USA

Volume

10

Issue

3

fYear

1995

fDate

6/1/1995 12:00:00 AM

Firstpage

16

Lastpage

22

Abstract

Investigates three potential neural network training algorithms in processing active sonar returns. Although all three methods generate reasonable probabilities of detection and false alarm in discriminating between man-made objects and background events, the stochastic training methods of simulated annealing and evolutionary programming outperform backpropagation

Keywords

backpropagation; genetic algorithms; learning (artificial intelligence); neural nets; probability; simulated annealing; sonar signal processing; stochastic processes; active sonar return processing; background events; backpropagation; detection probability; discriminating; evolutionary programming; false alarm probability; man-made objects; neural network training algorithms; simulated annealing; stochastic training methods; Genetic programming; Intelligent networks; Multilayer perceptrons; Neural networks; Object detection; Response surface methodology; Signal processing algorithms; Simulated annealing; Sonar; Stochastic processes;

fLanguage

English

Journal_Title

IEEE Expert

Publisher

ieee

ISSN

0885-9000

Type

jour

DOI

10.1109/64.393138

Filename

393138