Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published April 25, 2001 | Submitted
Report Open

The Central Classifier Bound - A New Error Bound for the Classifier Chosen by Early Stopping

Abstract

Training with early stopping is the following process. Partition the in sample data into training and validation sets Begin with a random classifier g_(1-). Use an iterative method to decrease the error rate on the training data. Record the classifier at each iteration producing a series of snapshots g_1....g_M. Evaluate the error rate of each snapshot over the validation data. Deliver a minimum validation error classifier. g^* as the result of training.

Additional Information

© 1997 California Institute of Technology. June 26, 1997. We thank Dr. Yaser Abu-Mostafa and Dr. Joel Franklin for their teaching and advice.

Attached Files

Submitted - CSTR1997.pdf

Submitted - postscript.ps

Files

CSTR1997.pdf
Files (481.3 kB)
Name Size Download all
md5:70d1bc98a249b628dfecbe9bd66fbe2b
321.5 kB Download
md5:9bd4ff1781c34fb37abdcecfa3979627
159.8 kB Preview Download

Additional details

Created:
August 19, 2023
Modified:
October 24, 2023