Translator Disclaimer
August 2005 Local Rademacher complexities
Peter L. Bartlett, Olivier Bousquet, Shahar Mendelson
Ann. Statist. 33(4): 1497-1537 (August 2005). DOI: 10.1214/009053605000000282

Abstract

We propose new bounds on the error of learning algorithms in terms of a data-dependent notion of complexity. The estimates we establish give optimal rates and are based on a local and empirical version of Rademacher averages, in the sense that the Rademacher averages are computed from the data, on a subset of functions with small empirical error. We present some applications to classification and prediction with convex function classes, and with kernel classes in particular.

Citation

Download Citation

Peter L. Bartlett. Olivier Bousquet. Shahar Mendelson. "Local Rademacher complexities." Ann. Statist. 33 (4) 1497 - 1537, August 2005. https://doi.org/10.1214/009053605000000282

Information

Published: August 2005
First available in Project Euclid: 5 August 2005

zbMATH: 1083.62034
MathSciNet: MR2166554
Digital Object Identifier: 10.1214/009053605000000282

Subjects:
Primary: 62G08, 68Q32

Rights: Copyright © 2005 Institute of Mathematical Statistics

JOURNAL ARTICLE
41 PAGES


SHARE
Vol.33 • No. 4 • August 2005
Back to Top