Inapproximability of Truthful Mechanisms via Generalizations of the VC Dimension.
Amit DanielyMichael SchapiraGal ShahafPublished in: CoRR (2014)
Keyphrases
- vc dimension
- mechanism design
- upper bound
- sample complexity
- lower bound
- distribution free
- concept classes
- covering numbers
- sample size
- vapnik chervonenkis dimension
- statistical learning theory
- empirical risk minimization
- inductive inference
- pac learnability
- pac learning
- concept class
- generalization bounds
- worst case
- approximation algorithms
- euclidean space
- theoretical analysis
- uniform convergence
- decision trees
- high dimensional
- np hard
- learning theory
- distance measure