Self-accelerated Thompson sampling with near-optimal regret upper bound.
Zhenyu ZhuLiusheng HuangHongli XuPublished in: Neurocomputing (2020)
Keyphrases
- upper bound
- lower bound
- worst case
- sample size
- branch and bound algorithm
- objective function
- branch and bound
- upper and lower bounds
- lower and upper bounds
- online learning
- regret bounds
- machine learning
- optimal solution
- np hard
- monte carlo
- sampling strategies
- random sampling
- sampling strategy
- sampling rate
- online algorithms
- multi agent systems
- minimax regret
- expert advice
- sampling algorithm
- generalization error
- special case
- learning problems
- model selection