Journal ArticleOptimal activation of halting multi‐armed bandit modelsCowan, Wesley [Computer Science Department Rutgers University Piscataway New Jersey USA]; Katehakis, Michael N [Management Science and Information Systems Department Rutgers University Piscataway New Jersey USA] (ORCID:0000000215117098); Ross, Sheldon M [Systems Engineering Department University of Southern California Los Angeles California USA]<title>Abstract</title> <p>We study new types of dynamic allocation problems the<italic>Halting Bandit</italic>models. As an application, we obtain new proofs for the classic Gittins index decomposition result compare Gittins (Journal of the Royal Statistical Society, Series B, 1979, 41, 148–177), and recent results of the authors in Cowan and Katehakis (Probability in the Engineering and Informational Sciences, 2015, 29, 51–76).</p>Wiley2023-10-0110669472Naval Research Logistics (NRL)707639 to 6520894-069Xhttps://doi.org/10.1002/nav.221451662629adaptive systems, autonomous reasoning and learning, dynamic data driven systems, machine learning, Markovian decision processesNational Science Foundation