Some Comparisons of Model Complexity in Linear and Neural-Network Approximation

0345940 - ÚI 2011 RIV DE eng C - Conference Paper (international conference)
Gnecco, G. - Kůrková, Věra - Sanguineti, M.
Some Comparisons of Model Complexity in Linear and Neural-Network Approximation.
Artificial Neural Networks – ICANN 2010. Vol. 3. Berlin: Springer, 2010 - (Diamantaras, K.; Duch, W.; Iliadis, L.), s. 358-367. Lecture Notes in Computer Science, 6354. ISBN 978-3-642-15824-7. ISSN 0302-9743.
[ICANN 2010. International Conference on Artificial Neural Networks /20./. Thessaloniki (GR), 15.09.2010-18.09.2010]
R&D Projects: GA MŠMT OC10047
Institutional research plan: CEZ:AV0Z10300504
Keywords : model complexity * neural networks * linear models
Subject RIV: IN - Informatics, Computer Science

Capabilities of linear and neural-network models are compared from the point of view of requirements on the growth of model complexity with an increasing accuracy of approximation. The bounds are formulated in terms of singular numbers of certain operators induced by computational units and high-dimensional volumes of the domains of the functions to be approximated.
Permanent Link: http://hdl.handle.net/11104/0187103