Learnability and the Vapnik-Chervonenkis dimension
Journal of the ACM (JACM)
k-Fold unions of low-dimensional concept classes
Information Processing Letters
Exploiting label dependencies for improved sample complexity
Machine Learning
Hi-index | 0.89 |
The known O(dklogk) bound on the VC dimension of k-fold unions or intersections of a given concept class with VC dimension d is shown to be asymptotically tight.