An Optimal Algorithm for Adversarial Bandits with Arbitrary Delays

We propose a new algorithm for adversarial multi-armed bandits with unrestricted delays. The algorithm is based on a novel hybrid regularizer applied in the Follow the Regularized Leader (FTRL) framework. It achieves regret guarantee, where is the number of arms, is the number of rounds, and is the total delay. The result matches the lower bound within constants and requires no prior knowledge of or . Additionally, we propose a refined tuning of the algorithm, which achieves regret guarantee, where is a set of rounds excluded from delay counting, are the counted rounds, and is the total delay in the counted rounds. If the delays are highly unbalanced, the latter regret guarantee can be significantly tighter than the former. The result requires no advance knowledge of the delays and resolves an open problem of Thune et al. (2019). The new FTRL algorithm and its refined tuning are anytime and require no doubling, which resolves another open problem of Thune et al. (2019).
View on arXiv