On using the Poincaré polynomial for calculating the VC dimension of neural networks.
Michael SchmittPublished in: Neural Networks (2001)
Keyphrases
- vc dimension
- vapnik chervonenkis dimension
- neural network
- upper bound
- sample complexity
- concept classes
- lower bound
- inductive inference
- sample size
- concept class
- covering numbers
- statistical learning theory
- compression scheme
- generalization bounds
- worst case
- pac learning
- uniform convergence
- special case
- euclidean space
- back propagation
- supervised learning
- statistical queries
- decision trees
- image processing