PAC Lower Bounds and Efficient Algorithms for The Max \(K\)-Armed Bandit Problem.
Yahel DavidNahum ShimkinPublished in: ICML (2016)
Keyphrases
- lower bound
- upper bound
- vc dimension
- sample complexity
- concept class
- statistical queries
- branch and bound algorithm
- branch and bound
- upper and lower bounds
- np hard
- objective function
- worst case
- pac learning
- lower and upper bounds
- sample size
- concept classes
- learning theory
- dnf formulas
- optimal solution
- lower bounding
- online algorithms
- pac learning model
- noise tolerant
- average case
- linear programming relaxation
- learning tasks
- multi class
- special case
- training data
- decision trees
- agnostic learning
- polynomial approximation
- learning algorithm