Radial Basis Function Neural Networks Have Superlinear VC Dimension.
Michael SchmittPublished in: COLT/EuroCOLT (2001)
Keyphrases
- vc dimension
- radial basis function neural network
- upper bound
- rbfnn
- pattern classification
- sample complexity
- sample size
- inductive inference
- radial basis function
- lower bound
- multilayer perceptron
- rbf neural network
- back propagation
- worst case
- sensitivity analysis
- neural network
- theoretical analysis
- euclidean space
- hidden layer
- generalization error
- special case
- training data
- text classification
- compression scheme
- knn
- rbf network
- artificial neural networks
- pattern recognition
- feature extraction