Abstract | ||
---|---|---|
We study the problem of when to stop learning a class of feedforward networks-- networks with linear outputs neuron and fixed input weights -- when they aretrained with a gradient descent algorithm on a finite number of examples. Undergeneral regularity conditions, it is shown that there are in general three distinctphases in the generalization performance in the learning process, and in particular,the network has better generalization performance when learning is stopped at acertain time ... |
Year | DOI | Venue |
---|---|---|
1993 | 10.1109/ISIT.1995.531518 | PROCEEDINGS 1995 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY |
Keywords | Field | DocType |
gradient descent,optimal stopping | Online machine learning,Early stopping,Mathematical optimization,Multi-task learning,Stability (learning theory),Probably approximately correct learning,Active learning (machine learning),Computer science,Wake-sleep algorithm,Artificial intelligence,Computational learning theory,Machine learning | Conference |
Citations | PageRank | References |
32 | 3.63 | 3 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wang, Changfeng | 1 | 32 | 3.63 |
Santosh S. Venkatesh | 2 | 381 | 71.80 |
J. Stephen Judd | 3 | 77 | 15.20 |