Related Experiment Videos
No free lunch for early stopping
Z Cataltepe1, Y S Abu-Mostafa, M Magdon-Ismail
1Bell Labs, Lucent Technologies, Room 2C-265, 600 Mountain Avenue, Murray Hill, NJ 07974, USA. zehra@bell-labs.com
Neural Computation
|May 5, 1999
Abstract:
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.