Fixed-Budget Best-Arm Identification in Contextual Bandits: A Static-Adaptive Algorithm.
Mohammad Javad AziziBranislav KvetonMohammad GhavamzadehPublished in: CoRR (2021)
Keyphrases
- experimental evaluation
- dynamic programming
- times faster
- detection algorithm
- learning algorithm
- k means
- cost function
- np hard
- similarity measure
- fixed size
- improved algorithm
- convex hull
- ant colony optimization
- computationally efficient
- expectation maximization
- high accuracy
- significant improvement
- search space
- search algorithm
- linear programming
- theoretical analysis
- optimization algorithm
- clustering method
- computational cost
- matching algorithm
- convergence rate
- estimation algorithm
- optimal solution
- identification rate