Interpolation networks

Author

Nunes, Luis ; Almeida, Luis B. ; Langlois, Thibault

Author_Institution

INESC, Lisbon, Portugal

Volume

3

fYear

1996

fDate

3-6 Jun 1996

Firstpage

1750

Abstract

This paper introduces a new type of network based on local response units, the `interpolation networks´ (INs). Under certain conditions this network is an interpolator. Its formulation allows a type of initialisation by prototypes that will set the net in a good initial starting point for the subsequent supervised learning process. These networks can be seen as a type of radial basis functions network (RBFN). However, their basis functions are essentially inverse squared distances instead of Gaussian functions. A brief description of the origin and motivation of INs is made in the first section, followed by the description of the first experiments with these networks

Keywords

feedforward neural nets; interpolation; learning (artificial intelligence); RBFN; initialisation; interpolation; interpolation networks; inverse squared distances; local response units; neural network; radial basis functions network; subsequent supervised learning process; Biological neural networks; Equations; Humans; Interpolation; Multilayer perceptrons; Nervous system; Prototypes; Radial basis function networks; Supervised learning; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1996., IEEE International Conference on

Conference_Location

Washington, DC

Print_ISBN

0-7803-3210-5

Type

conf

DOI

10.1109/ICNN.1996.549165

Filename

549165