Title :
An optimal interpolative net based pattern classifier
Author :
Eltoft, Torbjørn ; de Figueiredo, Rui J.P.
Author_Institution :
Inst. of Math. & Phys. Sci., Tromso Univ., Norway
fDate :
27 Jun-2 Jul 1994
Abstract :
In this paper we have developed a statistical pattern classifier, which applies the optimal interpolative network (OI-net), derived by one of the authors based on a generalized Fock space formulation. The classifier is a 3-stage algorithm, which uses an editing operation on the training set, and iteratively updates the net structure till the final solution is obtained. Its performance is demonstrated on two examples
Keywords :
interpolation; neural nets; optimisation; pattern classification; statistical analysis; 3-stage algorithm; generalized Fock space formulation; iterative updating; optimal interpolative neural-net-based pattern classifier; statistical pattern classifier; Contracts; Convergence; Ellipsoids; Iterative algorithms; Prototypes; Resonance light scattering; Signal analysis; Uncertainty;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374696
