Abstract
We present a method for determining the globally optimal on-line learning rule for a soft committee machine under a statistical mechanics framework. This rule maximizes the total reduction in generalization error over the whole learning process. A simple example demonstrates that the locally optimal rule, which maximizes the rate of decrease in generalization error, may perform poorly in comparison.
Original language | English |
---|---|
Pages (from-to) | L771-L776 |
Journal | Journal of Physics A: Mathematical and General |
Volume | 30 |
Issue number | 22 |
DOIs | |
Publication status | Published - 21 Nov 1997 |
Bibliographical note
Copyright of the Institute of Physics.Keywords
- globally optimal on-line learning
- soft committee machine
- error
- locally optimal rule