Cataltepe, Zehra and Abu-Mostafa, Yaser S. and Magdon-Ismail, Malik (1999) No Free Lunch for Early Stopping. Neural Computation, 11 (4). pp. 995-1009. ISSN 0899-7667. http://resolver.caltech.edu/CaltechAUTHORS:20111201-140641206
- Published Version
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20111201-140641206
We show that with a uniform prior on models having the same training error, early stopping at some fixed training error above the training error minimum results in an increase in the expected generalization error.
|Additional Information:||© 1999 Massachusetts Institute of Technology. Received January 20, 1998; accepted August 6, 1998. Posted Online March 13, 2006. We thank members of the Caltech Learning Systems Group; Amir Atiya, Alexander Nicholson, Joseph Sill, and Xubo Song, for many useful discussions, and two anonymous referees for their comments that improved the presentation of this article.|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Tony Diaz|
|Deposited On:||19 Jan 2012 21:47|
|Last Modified:||26 Dec 2012 14:33|
Repository Staff Only: item control page