Polynomial Bounds for VC Dimension of Sigmoidal and General Pfaffian Neural Networks.
Marek KarpinskiAngus MacintyrePublished in: J. Comput. Syst. Sci. (1997)
Keyphrases
- vc dimension
- vapnik chervonenkis dimension
- upper bound
- neural network
- pac learnability
- sample complexity
- lower bound
- vapnik chervonenkis
- sample size
- covering numbers
- inductive inference
- concept classes
- statistical learning theory
- distribution free
- generalization bounds
- special case
- empirical risk minimization
- learning machines
- worst case
- function classes
- pac learning
- euclidean space
- uniform convergence
- concept class
- compression scheme
- target function
- generalization error
- data compression
- artificial neural networks
- feature selection
- data sets
- risk bounds
- pac model
- learning problems