A VC-dimension-based Outer Bound on the Zero-Error Capacity of the Binary Adder Channel.
Or OrdentlichOfer ShayevitzPublished in: CoRR (2014)
Keyphrases
- vc dimension
- upper bound
- lower bound
- sample complexity
- distribution free
- vapnik chervonenkis
- generalization bounds
- sample size
- generalization error
- expected error
- statistical learning theory
- inductive inference
- concept classes
- vapnik chervonenkis dimension
- worst case
- covering numbers
- pac learning
- uniform convergence
- empirical risk minimization
- concept class
- learning machines
- euclidean space
- active learning
- compression scheme
- risk bounds
- learning algorithm
- function classes
- np hard
- data sets
- learning problems
- mind change complexity