Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Selecting ridge parameters in infinite dimensional hypothesis spaces

: Sugiyama, M.; Müller, K.-R.


Dorronsoro, J.R.:
ICANN 2002. International Conference on Artificial Neural Networks. Proceedings : Madrid, Spain, August 28 - 30, 2002
Berlin: Springer, 2002 (Lecture Notes in Computer Science 2415)
ISBN: 3-540-44074-7
ISSN: 0302-9743
International Conference on Artificial Neural Networks (ICANN) <12, 2002, Madrid>
Conference Paper
Fraunhofer FIRST ()

Previously, an unbiased estimator of the generalization error called the subspace information criterion (SIC) was proposed for a finite dimensional reproducing kernel Hilbert space (RKHS). In this paper, we extend SIC so that it can be applied to any RKHSs including infinite dimensional ones. Computer simulations show that the extended SIC works well in ridge parameter selection.