On the training of radial basis function classifiers
โ Scribed by M.T. Musavi; W. Ahmed; K.H. Chan; K.B. Faris; D.M. Hummels
- Publisher
- Elsevier Science
- Year
- 1992
- Tongue
- English
- Weight
- 671 KB
- Volume
- 5
- Category
- Article
- ISSN
- 0893-6080
No coin nor oath required. For personal study only.
โฆ Synopsis
An approach./or the implenwntation ql'the Radial Basis Function (RBF) technique is presented and applied to a network qlthe appropriate architecture. TIw paper evplores a melhodohlgy /br selecting kernel fimction parameters and tlw t:~tent to which the mtmber ~?/ RBF mules can be reduced without significantly q/feeling the overall training error These objectit'es are a~z'omplished through algorithms that shall be described ill detail. Emphasis is also placed on the problems faced h.l' a technique that has been proved stq~erior to the more traditional training algorithms, part iczthlr/)' ill terms o[processing.vJeed and soh,ahility q/'nonlinear patterns. Solutions are consequentl.l' proposed ill view o/making RBF a mrnv e[licient melhod./br interpolation and class(/icalion purposes.
๐ SIMILAR VOLUMES
Radial basis function interpolation has attracted a lot of interest in recent years. For popular choices, for example thin plate splines, this problem has a variational formulation, i.e. the interpolant minimizes a semi-norm on a certain space of radial functions. This gives rise to a function space