Local Rademacher complexities
Download (403.82 kB)
-
Altmetric Citations
Bartlett, Peter L.; Bousquet, Olivier; Mendelson, Shahar
Description
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.
Collections | ANU Research Publications |
---|---|
Date published: | 2005 |
Type: | Journal article |
URI: | http://hdl.handle.net/1885/97930 |
Source: | The Annals of Statistics |
DOI: | 10.1214/009053605000000282 |
Download
File | Description | Size | Format | Image |
---|---|---|---|---|
01_Bartlett_Local_Rademacher_c_2005.pdf | Published Version | 403.82 kB | Adobe PDF |
Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.
Updated: 17 November 2022/ Responsible Officer: University Librarian/ Page Contact: Library Systems & Web Coordinator