The Handbook of Brain Theory and Neural Networks. 2nd ed

0403755 - UIVT-O 20020140 RIV GB eng M - Část monografie knihy
Kůrková, Věra
Universal Approximators.
The Handbook of Brain Theory and Neural Networks. 2nd ed. Vol. 3. Cambridge: The MIT Press, 2002 - (Arbib, M.), s. 1180-1183. ISBN 0-262-01197-2
Grant CEP: GA ČR GA201/99/0092; GA ČR GA201/02/0428
Klíčová slova: universal approximation property * rates of approximation * curse of dimensionality
Kód oboru RIV: BA - Obecná matematika

This article surveys recent developments in the mathematical theory of feedforward networks and includes proofs of the universal approximation capabilities of perceptron and RBF networks with gengeral activation and radial functions, estimates of rates of approximation and description of sets of multivariable functions that can be approximated without the "curse of dimensionality".
Trvalý link: http://hdl.handle.net/11104/0124047