The VC-Dimension vs. the Statistical Capacity for Two Layer Networks with Binary Weights.
Chuanyi JiDemetri PsaltisPublished in: COLT (1991)
Keyphrases
- vc dimension
- vapnik chervonenkis dimension
- upper bound
- concept classes
- sample complexity
- sample size
- covering numbers
- empirical risk minimization
- lower bound
- statistical learning theory
- distribution free
- concept class
- pac learnability
- inductive inference
- compression scheme
- learning theory
- worst case
- euclidean space
- pac learning
- supervised learning
- special case
- multiresolution
- image processing