Dual-Directed Algorithm Design for Efficient Pure Exploration

The authors thank the anonymous reviewers and associate editor for valuable feedback that significantly improved the paper and Daniel Russo, Sandeep Juneja, Po-An Wang, Shane Henderson, Xiaowei Zhang, Jun Luo, the participants of the 2023 INFORMS Applied Probability Society Conference, the 2023 INFORMS Annual Meeting, and the 2024 INFORMS MSOM Conference for their insightful comments on this work. A preliminary version of this paper appeared as an extended abstract in the Proceedings of the 36th Annual Conference on Learning Theory, with the title “Information-Directed Selection for Top-Two Algorithms.”