MEASURING THE VC-DIMENSION OF A LEARNING-MACHINE

Authors
VAPNIK V LEVIN E LECUN Y
Citation
V. Vapnik et al., MEASURING THE VC-DIMENSION OF A LEARNING-MACHINE, Neural computation, 6(5), 1994, pp. 851-876
Citations number
10
Categorie Soggetti
Computer Sciences","Computer Science Artificial Intelligence",Neurosciences
Journal title
Neural computation → ACNP
ISSN journal
08997667
Volume
6
Issue
5
Year of publication
1994
Pages
851 - 876
Database
ISI
SICI code
0899-7667(1994)6:5<851:MTVOAL>2.0.ZU;2-F
Abstract
A method for measuring the capacity of learning machines is described. The method is based on fitting a theoretically derived function to em pirical measurements of the maximal difference between the error rates on two separate data sets of varying sizes. Experimental measurements of the capacity of various types of linear classifiers are presented.