A self-organizing neural network for multidimensional approximation

Author

Palmieri, Francesco

Author_Institution

Dept. of Electr. & Syst. Eng., Connecticut Univ., Storrs, CT, USA

Volume

4

fYear

1992

fDate

7-11 Jun 1992

Firstpage

802

Abstract

A neural network based on the combination of a feature map (memory) and linear filters is proposed as a generalized adaptive processor for multidimensional nonlinear mapping. The self-organizing part of the system provides a progressively finer embedding of the input space as more units are added to the network. The linear filters, which tap from the memory, provide the function approximations. Learning is achieved with simple rules of the Hebb´s type with no backpropagation needed. The author reports some preliminary results on two-dimensional patterns that show the potential of this approach

Keywords

function approximation; pattern recognition; self-organising feature maps; Hebbian learning; feature map; function approximations; generalized adaptive processor; linear filters; multidimensional approximation; nonlinear mapping; self-organizing neural network; two-dimensional patterns; Backpropagation; Biological neural networks; Filtering; Function approximation; Least squares approximation; Multidimensional systems; Neural networks; Nonlinear filters; Size control; Systems engineering and theory;

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.1992.227219

Filename

227219