Published April 30, 2001
| Submitted
Report
Open
No Free Lunch for Early Stopping
Abstract
We show that, with a uniform prior on hypothesis functions 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. We also show that regularization methods are equivalent to early stopping with certain non-uniform prior on the early stopping solutions.
Additional Information
© 1998 Zehra Cataltepe, California Institute of Technology. January 19, 1998. We would like to thank members of the Caltech Learning Systems Group Dr Amir Atiya, Alexander Nicholson, Joseph Sill and Xubo Song for many useful discussions.Attached Files
Submitted - CSTR1998.pdf
Submitted - postscript.ps
Files
CSTR1998.pdf
Files
(634.2 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:992a53d59b2fb15450798ceb7d79b393
|
435.2 kB | Download |
|
md5:1bcf586cf8d8ab6c85391c69cd0dd424
|
199.0 kB | Preview Download |
Additional details
- Eprint ID
- 26836
- Resolver ID
- CaltechCSTR:1998.cs-tr-98-02
- Created
-
2001-04-30Created from EPrint's datestamp field
- Updated
-
2019-10-03Created from EPrint's last_modified field
- Caltech groups
- Computer Science Technical Reports