On the generalization ability of GRLVQ networks
B Hammer, M Strickert, T Villmann - Neural Processing Letters, 2005 - Springer
B Hammer, M Strickert, T Villmann
Neural Processing Letters, 2005•SpringerWe derive a generalization bound for prototype-based classifiers with adaptive metric. The
bound depends on the margin of the classifier and is independent of the dimensionality of
the data. It holds for classifiers based on the Euclidean metric extended by adaptive
relevance terms. In particular, the result holds for relevance learning vector quantization
(RLVQ)[4] and generalized relevance learning vector quantization (GRLVQ)[19].
bound depends on the margin of the classifier and is independent of the dimensionality of
the data. It holds for classifiers based on the Euclidean metric extended by adaptive
relevance terms. In particular, the result holds for relevance learning vector quantization
(RLVQ)[4] and generalized relevance learning vector quantization (GRLVQ)[19].
Abstract
We derive a generalization bound for prototype-based classifiers with adaptive metric. The bound depends on the margin of the classifier and is independent of the dimensionality of the data. It holds for classifiers based on the Euclidean metric extended by adaptive relevance terms. In particular, the result holds for relevance learning vector quantization (RLVQ) [4] and generalized relevance learning vector quantization (GRLVQ) [19].
Springer
Showing the best result for this search. See all results