ML Anthology

Localized Rademacher Complexities

Peter L. Bartlett, Olivier Bousquet, Shahar Mendelson
COLT 2002 pp. 44-58
doi:10.1007/3-540-45435-7_4 /colt/2002/bartlett2002colt-localized/

Abstract

We investigate the behaviour of global and local Rademacher averages. We present new error bounds which are based on the local averages and indicate how data-dependent local averages can be estimated without a priori knowledge of the class at hand.

PDF COLT Semantic Scholar

Cite

Text

Bartlett et al. "Localized Rademacher Complexities." Annual Conference on Computational Learning Theory, 2002. doi:10.1007/3-540-45435-7_4

Markdown

[Bartlett et al. "Localized Rademacher Complexities." Annual Conference on Computational Learning Theory, 2002.](https://mlanthology.org/colt/2002/bartlett2002colt-localized/) doi:10.1007/3-540-45435-7_4

BibTeX

@inproceedings{bartlett2002colt-localized,
  title     = {{Localized Rademacher Complexities}},
  author    = {Bartlett, Peter L. and Bousquet, Olivier and Mendelson, Shahar},
  booktitle = {Annual Conference on Computational Learning Theory},
  year      = {2002},
  pages     = {44-58},
  doi       = {10.1007/3-540-45435-7_4},
  url       = {https://mlanthology.org/colt/2002/bartlett2002colt-localized/}
}