In this paper we compare variants of elliptical basis function networks for classification tasks. The networks are introduced as density estimators and then modified towards RBF networks. Node reduction is accomplished by a genetic algorithm(GA). Two different kinds of node connections are compared. As a second degree of freedom different types of basis functions are investigated. On an artificial test set of the time series domain the impact of dimensionality is considered.
CITATION STYLE
Gutjahr, S., & Feist, J. (1997). Elliptical basis function networks for classification tasks. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1327, pp. 373–378). Springer Verlag. https://doi.org/10.1007/bfb0020183
Mendeley helps you to discover research relevant for your work.