An Optimal Algorithm for Adversarial Bandits with Arbitrary Delays.
Julian ZimmertYevgeny SeldinPublished in: CoRR (2019)
Keyphrases
- optimal solution
- dynamic programming
- worst case
- experimental evaluation
- detection algorithm
- improved algorithm
- learning algorithm
- computational cost
- preprocessing
- significant improvement
- np hard
- path planning
- tree structure
- theoretical analysis
- times faster
- globally optimal
- closed form
- matching algorithm
- locally optimal
- optimization algorithm
- segmentation algorithm
- expectation maximization
- classification algorithm
- linear programming
- upper bound
- probabilistic model
- cost function
- convergence rate
- computational complexity
- similarity measure
- genetic algorithm