Polynomial Bounds for VC Dimension of Sigmoidal Neural Networks
Marek KarpinskiAngus MacintyrePublished in: Electron. Colloquium Comput. Complex. (1994)
Keyphrases
- vc dimension
- neural network
- vapnik chervonenkis dimension
- upper bound
- vapnik chervonenkis
- covering numbers
- lower bound
- sample complexity
- sample size
- generalization bounds
- distribution free
- inductive inference
- learning machines
- empirical risk minimization
- concept classes
- statistical learning theory
- worst case
- compression scheme
- function classes
- concept class
- euclidean space
- uniform convergence
- upper and lower bounds
- target function
- back propagation
- large deviations
- dnf formulas
- pac learning
- special case
- artificial neural networks
- risk bounds
- training data
- learning algorithm