Model Selection for Support Vector Machines

Part of Advances in Neural Information Processing Systems 12 (NIPS 1999)

Bibtex Metadata Paper


Olivier Chapelle, Vladimir Vapnik


New functionals for parameter (model) selection of Support Vector Ma(cid:173) chines are introduced based on the concepts of the span of support vec(cid:173) tors and rescaling of the feature space. It is shown that using these func(cid:173) tionals, one can both predict the best choice of parameters of the model and the relative quality of performance for any value of parameter.