| \begin{thebibliography}{26} |
| \providecommand{\natexlab}[1]{#1} |
| \providecommand{\url}[1]{\texttt{#1}} |
| \expandafter\ifx\csname urlstyle\endcsname\relax |
| \providecommand{\doi}[1]{doi: |
| \providecommand{\doi}{doi: \begingroup \urlstyle{rm}\Url}\fi |
|
|
| \bibitem[Agrawal(2023)]{agrawal2022bandits} |
| Agrawal, S. |
| \newblock \emph{Bandits with Heavy Tails: Algorithms Analysis and Optimality}. |
| \newblock PhD thesis, Tata Institute of Fundamental Research, 2023. |
| \newblock URL \url{http://hdl.handle.net/10603/478863}. |
|
|
| \bibitem[Agrawal \& Ramdas(2025)Agrawal and Ramdas]{agrawal2025stopping} |
| Agrawal, S. and Ramdas, A. |
| \newblock On stopping times of power-one sequential tests: Tight lower and upper bounds. |
| \newblock \emph{arXiv preprint arXiv:2504.19952}, 2025. |
|
|
| \bibitem[Agrawal et~al.(2020)Agrawal, Juneja, and Glynn]{agrawal2020optimal} |
| Agrawal, S., Juneja, S., and Glynn, P. |
| \newblock Optimal $\delta$-correct best-arm selection for heavy-tailed distributions. |
| \newblock In \emph{Algorithmic Learning Theory}, pp.\ 61--110. PMLR, 2020. |
|
|
| \bibitem[Agrawal et~al.(2021{\natexlab{a}})Agrawal, Juneja, and Koolen]{pmlr-v134-agrawal21a} |
| Agrawal, S., Juneja, S.~K., and Koolen, W.~M. |
| \newblock Regret minimization in heavy-tailed bandits. |
| \newblock In \emph{Proceedings of Thirty Fourth Conference on Learning Theory}, volume 134 of \emph{Proceedings of Machine Learning Research}, pp.\ 26--62. PMLR, 15--19 Aug 2021{\natexlab{a}}. |
|
|
| \bibitem[Agrawal et~al.(2021{\natexlab{b}})Agrawal, Koolen, and Juneja]{agrawal2021optimal} |
| Agrawal, S., Koolen, W.~M., and Juneja, S. |
| \newblock Optimal best-arm identification methods for tail-risk measures. |
| \newblock \emph{Advances in Neural Information Processing Systems}, 34:\penalty0 25578--25590, 2021{\natexlab{b}}. |
|
|
| \bibitem[Anscombe(1952)]{anscombe1952large} |
| Anscombe, F.~J. |
| \newblock Large-sample theory of sequential estimation. |
| \newblock In \emph{Mathematical Proceedings of the Cambridge Philosophical Society}, volume~48, pp.\ 600--607. Cambridge University Press, 1952. |
|
|
| \bibitem[Asmussen(2003)]{asmussen2003applied} |
| Asmussen, S. |
| \newblock \emph{Applied probability and queues}. |
| \newblock Springer, 2003. |
|
|
| \bibitem[Billingsley(2017)]{billingsley2017probability} |
| Billingsley, P. |
| \newblock \emph{Probability and measure}. |
| \newblock John Wiley \& Sons, 2017. |
|
|
| \bibitem[Burnetas \& Katehakis(1996)Burnetas and Katehakis]{burnetas1996optimal} |
| Burnetas, A.~N. and Katehakis, M.~N. |
| \newblock Optimal adaptive policies for sequential allocation problems. |
| \newblock \emph{Advances in Applied Mathematics}, 17\penalty0 (2):\penalty0 122--142, 1996. |
|
|
| \bibitem[Chernoff(1959)]{chernoff1992sequential} |
| Chernoff, H. |
| \newblock Sequential design of experiments. |
| \newblock \emph{Ann. Math. Statist.}, 30\penalty0 (4):\penalty0 755--770, 1959. |
|
|
| \bibitem[Darling \& Robbins(1967)Darling and Robbins]{darling1967iterated} |
| Darling, D.~A. and Robbins, H. |
| \newblock Iterated logarithm inequalities. |
| \newblock \emph{Proceedings of the National Academy of Sciences}, 57\penalty0 (5):\penalty0 1188--1192, 1967. |
|
|
| \bibitem[Deep et~al.(2024)Deep, Bassamboo, and Juneja]{deep2024asymptotically} |
| Deep, V., Bassamboo, A., and Juneja, S.~K. |
| \newblock Asymptotically optimal and computationally efficient average treatment effect estimation in a/b testing. |
| \newblock 2024. |
|
|
| \bibitem[Deep et~al.(2025)Deep, Bassamboo, and Juneja]{deep2025asymptotic} |
| Deep, V., Bassamboo, A., and Juneja, S. |
| \newblock Asymptotic optimality theory of confidence intervals of the mean. |
| \newblock \emph{arXiv preprint arXiv:2501.19126}, 2025. |
|
|
| \bibitem[Fan \& Glynn(2025)Fan and Glynn]{fan2025fragility} |
| Fan, L. and Glynn, P.~W. |
| \newblock The fragility of optimized bandit algorithms. |
| \newblock \emph{Operations Research}, 73\penalty0 (6):\penalty0 3173--3198, 2025. |
|
|
| \bibitem[Gut(2009)]{gut2009stopped} |
| Gut, A. |
| \newblock \emph{Stopped random walks}. |
| \newblock Springer, 2009. |
|
|
| \bibitem[Honda \& Takemura(2010)Honda and Takemura]{honda2010asymptotically} |
| Honda, J. and Takemura, A. |
| \newblock An asymptotically optimal bandit algorithm for bounded support models. |
| \newblock In \emph{COLT}, pp.\ 67--79, 2010. |
|
|
| \bibitem[Honda \& Takemura(2015)Honda and Takemura]{honda2015non} |
| Honda, J. and Takemura, A. |
| \newblock Non-asymptotic analysis of a new bandit algorithm for semi-bounded rewards. |
| \newblock \emph{J. Mach. Learn. Res.}, 16:\penalty0 3721--3756, 2015. |
|
|
| \bibitem[Jourdan et~al.(2022)Jourdan, Degenne, Baudry, de~Heide, and Kaufmann]{jourdan2022top} |
| Jourdan, M., Degenne, R., Baudry, D., de~Heide, R., and Kaufmann, E. |
| \newblock Top two algorithms revisited. |
| \newblock \emph{Advances in Neural Information Processing Systems}, 35:\penalty0 26791--26803, 2022. |
|
|
| \bibitem[Lai \& Robbins(1985)Lai and Robbins]{lai1985asymptotically} |
| Lai, T.~L. and Robbins, H. |
| \newblock Asymptotically efficient adaptive allocation rules. |
| \newblock \emph{Advances in Applied Mathematics}, 6\penalty0 (1):\penalty0 4--22, 1985. |
|
|
| \bibitem[Mukhopadhyay(2020)]{mukhopadhyay2020asymptotic} |
| Mukhopadhyay, N. |
| \newblock Asymptotic normality of sequential stopping times with applications: Confidence intervals for an exponential mean. |
| \newblock \emph{Calcutta Statistical Association Bulletin}, 72\penalty0 (1):\penalty0 17--34, 2020. |
|
|
| \bibitem[Panda \& Agrawal(2026)Panda and Agrawal]{panda2026regret} |
| Panda, S. and Agrawal, S. |
| \newblock Regret tail characterization of optimal bandit algorithms with generic rewards. |
| \newblock \emph{arXiv preprint arXiv:2604.14876}, 2026. |
|
|
| \bibitem[Robbins \& Siegmund(1974)Robbins and Siegmund]{robbins1974expected} |
| Robbins, H. and Siegmund, D. |
| \newblock The expected sample size of some tests of power one. |
| \newblock \emph{The Annals of Statistics}, 2\penalty0 (3):\penalty0 415--436, 1974. |
|
|
| \bibitem[Siegmund(2013)]{siegmund2013sequential} |
| Siegmund, D. |
| \newblock \emph{Sequential Analysis: Tests and Confidence Intervals}. |
| \newblock Springer Science \& Business Media, 2013. |
|
|
| \bibitem[Wald(1992)]{wald1992sequential} |
| Wald, A. |
| \newblock Sequential tests of statistical hypotheses. |
| \newblock In \emph{Breakthroughs in statistics: Foundations and basic theory}, pp.\ 256--298. Springer, 1992. |
|
|
| \bibitem[Wald \& Wolfowitz(1948)Wald and Wolfowitz]{wald1948optimum} |
| Wald, A. and Wolfowitz, J. |
| \newblock Optimum character of the sequential probability ratio test. |
| \newblock \emph{The Annals of Mathematical Statistics}, pp.\ 326--339, 1948. |
|
|
| \bibitem[Wang et~al.(2026)Wang, Agrawal, and Ramdas]{wang2026almost} |
| Wang, H., Agrawal, S., and Ramdas, A. |
| \newblock Almost sure null bankruptcy of testing-by-betting strategies. |
| \newblock \emph{arXiv preprint arXiv:2602.08888}, 2026. |
|
|
| \end{thebibliography} |
|
|