Rademacher Complexities and Bounding the Excess Risk in Active Learning
Abstract
Sequential algorithms of active learning based on the estimation of the level sets of the empirical risk are discussed in the paper. Localized Rademacher complexities are used in the algorithms to estimate the sample sizes needed to achieve the required accuracy of learning in an adaptive way. Probabilistic bounds on the number of active examples have been proved and several applications to binary classification problems are considered.
Cite
Text
Koltchinskii. "Rademacher Complexities and Bounding the Excess Risk in Active Learning." Journal of Machine Learning Research, 2010.Markdown
[Koltchinskii. "Rademacher Complexities and Bounding the Excess Risk in Active Learning." Journal of Machine Learning Research, 2010.](https://mlanthology.org/jmlr/2010/koltchinskii2010jmlr-rademacher/)BibTeX
@article{koltchinskii2010jmlr-rademacher,
title = {{Rademacher Complexities and Bounding the Excess Risk in Active Learning}},
author = {Koltchinskii, Vladimir},
journal = {Journal of Machine Learning Research},
year = {2010},
pages = {2457-2485},
volume = {11},
url = {https://mlanthology.org/jmlr/2010/koltchinskii2010jmlr-rademacher/}
}