Improved risk tail bounds for on-line algorithms
Nicolò Cesa-Bianchi and Claudio Gentile
In: NIPS 2005, 5-8 Dec 2005, Vancouver, Canada.

There is a more recent version of this eprint available. Click here to view it.

Abstract

We prove the strongest known bound for the risk of hypotheses selected from the ensemble generated by running a learning algorithm incrementally on the training data. Our result is based on proof techniques that are remarkably different from the standard risk analysis based on uniform convergence arguments.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Available Versions of this Item

• Improved risk tail bounds for on-line algorithms (deposited 28 November 2005) [Currently Displayed]
  • Improved risk tail bounds for on-line algorithms (deposited 14 February 2008)

[Edit]