Surrogate solutions of Fredholm equations by feedforward networks

0384883 - ÚI 2013 RIV SK eng C - Konferenční příspěvek (zahraniční konf.)
Kůrková, Věra
Surrogate solutions of Fredholm equations by feedforward networks.
Information Technologies - Applications and Theory. Seňa: PONT s.r.o., 2012 - (Horváth, T.), s. 49-54. ISBN 978-80-971144-0-4.
[ITAT 2012. Conference on Theory and Practice of Information Technologies. Ždiar (SK), 17.09.2012-21.09.2012]
Grant CEP: GA MŠMT OC10047
Institucionální podpora: RVO:67985807
Klíčová slova: surrogate modeling by neural networks * Fredholm integral equations * rates of approximation * model complexity
Kód oboru RIV: IN - Informatika
http://ceur-ws.org/Vol-990/paper8.pdf

Surrogate solutions of Fredholm integral equations by feedforward neural networks are investigated theoretically. Convergence of surrogate solutions computable by networks with increasing numbers of computational units to theoretically optimal solutions is proven and upper bounds on rates of convergence are derived. The results hold for a variety of computational units, they are illustrated by examples of perceptrons and Gaussian radial units.
Trvalý link: http://hdl.handle.net/11104/0007335