Surrogate Modelling of Solutions of Integral Equations by Neural Networks

0379775 - ÚI 2013 RIV DE eng C - Konferenční příspěvek (zahraniční konf.)
Kůrková, Věra
Surrogate Modelling of Solutions of Integral Equations by Neural Networks.
Artificial Intelligence Applications and Innovations. Heidelberg: Springer, 2012 - (Iliadis, L.; Maglogiannis, I.; Papadopoulos, H.), s. 88-96. IFIP Advances in Information and Communication Technology, 381. ISBN 978-3-642-33408-5. ISSN 1868-4238.
[AIAI 2012. IFIP WG 12.5 International Conference /8./. Halkidiki (GR), 27.09.2012-30.09.2012]
Grant CEP: GA MŠMT OC10047
Institucionální podpora: RVO:67985807
Klíčová slova: surrogate modelling by neural networks * Fredholm integral equations * rates of approximation * model complexity
Kód oboru RIV: IN - Informatika

Surrogate modelling of solutions of integral equations by neural networks is investigated theoretically. Estimates of speed of convergence of suboptimal surrogate solutions to solutions described by Fredholm theorem are derived.
Trvalý link: http://hdl.handle.net/11104/0210655