Gaussian Radial and Kernel Networks with Varying and Fixed Widths

0389193 - ÚI 2013 RIV CZ eng C - Konferenční příspěvek (zahraniční konf.)
Kůrková, Věra
Gaussian Radial and Kernel Networks with Varying and Fixed Widths.
SOFSEM 2013: Theory and Practice of Computer Science. Vol. 2. Prague: Institute of Computer Science AS CR, 2013 - (van Emde Boas, P.; Groen, F.; Italiano, G.; Nawrocki, J.; Sack, H.), s. 95-102. ISBN 978-80-87136-15-7.
[SOFSEM 2013. Conference on Current Trends in Theory and Practice of Computer Science /39./. Špindlerův Mlýn (CZ), 26.01.2013-31.01.2013]
Grant CEP: GA ČR GAP202/11/1368
Institucionální podpora: RVO:67985807
Klíčová slova: Gaussian radial and kernel networks * functionally equivalent networks * universal approximators * stabilizers defined by Gaussian kernels
Kód oboru RIV: IN - Informatika

The role of widths of Gaussians in computational models which they generate is investigated. Suitability of Gaussian kernel models with fixed widths for regression is proven in terms of their universal approximation capability. Large sets of argminima of error functionals minimized during learning from data over Gaussian networks with varying widths are described. Dependence of stabilizers modelling generalization on widths of Gaussian kernels and the input dimension is estimated.
Trvalý link: http://hdl.handle.net/11104/0218073