ML Anthology

Fast Rates to Bayes for Kernel Machines

Ingo Steinwart, Clint Scovel
NeurIPS 2004 pp. 1345-1352
/neurips/2004/steinwart2004neurips-fast/

Abstract

We establish learning rates to the Bayes risk for support vector machines (SVMs) with hinge loss. In particular, for SVMs with Gaussian RBF kernels we propose a geometric condition for distributions which can be used to determine approximation properties of these kernels. Finally, we compare our methods with a recent paper of G. Blanchard et al..

PDF NeurIPS Semantic Scholar

Cite

Text

Steinwart and Scovel. "Fast Rates to Bayes for Kernel Machines." Neural Information Processing Systems, 2004.

Markdown

[Steinwart and Scovel. "Fast Rates to Bayes for Kernel Machines." Neural Information Processing Systems, 2004.](https://mlanthology.org/neurips/2004/steinwart2004neurips-fast/)

BibTeX

@inproceedings{steinwart2004neurips-fast,
  title     = {{Fast Rates to Bayes for Kernel Machines}},
  author    = {Steinwart, Ingo and Scovel, Clint},
  booktitle = {Neural Information Processing Systems},
  year      = {2004},
  pages     = {1345-1352},
  url       = {https://mlanthology.org/neurips/2004/steinwart2004neurips-fast/}
}