DocumentCode :
2363152
Title :
A self-organizing system for the development of neural network parameter estimators
Author :
Manry, M.T.
Author_Institution :
Dept. of Electr. Eng., Texas Univ., Arlington, TX, USA
fYear :
1995
fDate :
31 Aug-2 Sep 1995
Firstpage :
105
Lastpage :
114
Abstract :
The design an optimal neural network estimator from training data is difficult because: 1) the required complexity of the estimation network is unknown, 2) existing training algorithms for multilayer perceptrons (MLPs) are inefficient, in terms of training time and use of free parameters, 3) existing bounds on neural network estimation error assume noiseless inputs and are not practical to calculate, 4) there is no generally accepted procedure for finding the best subset of input features to be used in optimal estimation, and 5) a method for automatically developing optimal estimators from training data is not available. In this paper, we present a methodology for attacking these problems. We describe three separate processing blocks which attempt to solve problems (1), (2), and (3). These blocks are then assembled into larger compound systems or blocks which attempt to solve the remaining problems. Examples of multilayer perceptron (MLP) estimators, designed using the proposed system, are given
Keywords :
feedforward neural nets; learning (artificial intelligence); multilayer perceptrons; parameter estimation; self-organising feature maps; complexity estimation; learning algorithm; multilayer perceptrons; neural network; parameter estimators; processing blocks; self-organizing system; Backpropagation algorithms; Character generation; Clustering algorithms; Filters; Multi-layer neural network; Multilayer perceptrons; Neural networks; Parameter estimation; Polynomials; Training data;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks for Signal Processing [1995] V. Proceedings of the 1995 IEEE Workshop
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-2739-X
Type :
conf
DOI :
10.1109/NNSP.1995.514884
Filename :
514884
Link To Document :
بازگشت