A VC-dimension-based outer bound on the zero-error capacity of the binary adder channel.
Or OrdentlichOfer ShayevitzPublished in: ISIT (2015)
Keyphrases
- vc dimension
- upper bound
- sample complexity
- lower bound
- vapnik chervonenkis
- distribution free
- sample size
- concept classes
- generalization bounds
- statistical learning theory
- generalization error
- covering numbers
- expected error
- vapnik chervonenkis dimension
- worst case
- inductive inference
- empirical risk minimization
- uniform convergence
- concept class
- pac learning
- learning machines
- compression scheme
- euclidean space
- theoretical analysis
- special case
- training data
- machine learning
- objective function
- image compression