A PAC algorithm in relative precision for bandit problem with costly sampling.
Marie Billaud FriessArthur MachereyAnthony NouyClémentine PrieurPublished in: CoRR (2020)
Keyphrases
- learning algorithm
- times faster
- preprocessing
- matching algorithm
- sampling algorithm
- objective function
- recognition algorithm
- computational complexity
- experimental evaluation
- detection algorithm
- expectation maximization
- worst case
- dynamic programming
- k means
- linear programming
- particle swarm optimization
- cost function
- significant improvement
- optimal solution
- similarity measure
- computational cost
- np hard
- optimization algorithm
- evolutionary algorithm