Some Comparisons of Networks with Radial and Kernel Units

0379010 - ÚI 2013 RIV DE eng C - Konferenční příspěvek (zahraniční konf.)
Kůrková, Věra
Some Comparisons of Networks with Radial and Kernel Units.
Artificial Neural Networks and Machine Learning – ICANN 2012. Berlin: Springer, 2012 - (Villa, A.; Duch, W.; Érdi, P.; Masulli, F.; Palm, G.), s. 17-24. Lecture Notes in Computer Science, 7553. ISBN 978-3-642-33265-4. ISSN 0302-9743.
[ICANN 2012. International Conference on Artificial Neural Networks /22./. Lausanne (CH), 11.09.2012-14.09.2012]
Grant CEP: GA ČR GAP202/11/1368
Institucionální podpora: RVO:67985807
Klíčová slova: radial and kernel networks * universal approximation property * fixed and varying widths * minimization of error functionals * stabilizers induced by kernels
Kód oboru RIV: IN - Informatika

Two types of computational models, radial-basis function networks with units having varying widths and kernel networks where all units have a fixed width, are investigated in the framework of scaled kernels. The impact of widths of kernels on approximation of multivariable functions, generalization modelled by regularization with kernel stabilizers, and minimization of error functionals is analyzed.
Trvalý link: http://hdl.handle.net/11104/0210297