Almost Optimal Anytime Algorithm for Batched Multi-Armed Bandits
Tianyuan Jin 1 Jing Tang 2 Pan Xu 3 Keke Huang 1 Xiaokui Xiao 1 Quanquan Gu 3
Abstract it to guide the next action. However, this is impractical for
many real applications where the number of interactions
In batched multi-armed bandit problems, the between the agent and environment is limited. For example,
learner can adaptively pul ...


雷达卡




京公网安备 11010802022788号







