Inapproximability of Truthful Mechanisms via Generalizations of the VC Dimension.
Amit DanielyMichael SchapiraGal ShahafPublished in: STOC (2015)
Keyphrases
- vc dimension
- mechanism design
- upper bound
- sample size
- sample complexity
- lower bound
- inductive inference
- empirical risk minimization
- concept classes
- vapnik chervonenkis dimension
- approximation algorithms
- statistical learning theory
- pac learnability
- concept class
- worst case
- covering numbers
- distribution free
- generalization bounds
- uniform convergence
- pac learning
- compression scheme
- euclidean space
- function classes
- np hard
- data sets