Nearly-tight VC-dimension bounds for piecewise linear neural networks.
Nick HarveyChristopher LiawAbbas MehrabianPublished in: CoRR (2017)
Keyphrases
- vc dimension
- piecewise linear
- upper bound
- lower bound
- neural network
- worst case
- vapnik chervonenkis
- covering numbers
- generalization bounds
- sample complexity
- concept classes
- sample size
- distribution free
- statistical learning theory
- empirical risk minimization
- dynamic programming
- uniform convergence
- concept class
- inductive inference
- pattern recognition
- hyperplane
- optimal solution
- generalization error
- function classes
- data sets
- special case
- objective function
- learning problems
- training samples
- nearest neighbor
- active learning
- np hard