Vapnik-Chervonenkis bounds for generalization
- 7 May 1993
- journal article
- Published by IOP Publishing in Journal of Physics A: General Physics
- Vol. 26 (9) , 2211-2223
- https://doi.org/10.1088/0305-4470/26/9/016
Abstract
The authors review the Vapnik and Chervonenkis theorem as applied to the problem of generalization. By combining some of the technical modifications proposed in the literature they derive tighter bounds and a new version of the theorem bounding the accuracy in the estimation of generalization probabilities from finite samples. A critical discussion and comparison with the results from statistical mechanics is given.Keywords
This publication has 18 references indexed in Scilit:
- Statistical mechanics of learning from examplesPhysical Review A, 1992
- Learnability and the Vapnik-Chervonenkis dimensionJournal of the ACM, 1989
- The Vapnik-Chervonenkis Dimension: Information versus Complexity in LearningNeural Computation, 1989
- Three unfinished works on the optimal storage capacity of networksJournal of Physics A: General Physics, 1989
- What Size Net Gives Valid Generalization?Neural Computation, 1989
- A theory of the learnableCommunications of the ACM, 1984
- Bounds for the uniform deviation of empirical measuresJournal of Multivariate Analysis, 1982
- On the density of families of setsJournal of Combinatorial Theory, Series A, 1972
- On the Uniform Convergence of Relative Frequencies of Events to Their ProbabilitiesTheory of Probability and Its Applications, 1971
- On the Distribution of the Number of Successes in Independent TrialsThe Annals of Mathematical Statistics, 1956