Rademacher processes and bounding the risk of function learning
Probability
2007-05-23 v1
Abstract
We construct data dependent bounds on the risk in function learning problems. The bounds are based on the local norms of the Rademacher process indexed by the underlying function class and they do not require prior knowledge about the distribution of the training examples or any specific properties of the function class.
Cite
@article{arxiv.math/0405338,
title = {Rademacher processes and bounding the risk of function learning},
author = {Vladimir Koltchinskii and Dmitry Panchenko},
journal= {arXiv preprint arXiv:math/0405338},
year = {2007}
}
Comments
14 pages