Adaptive control of discounted Markov decision chains (Q796461)

We consider discounted-reward finite-state Markov decision processes which depend on unknown parameters. An adaptive policy inspired by the nonstationary value iteration scheme of \textit{A. Federgruen} and \textit{P. J. Schweitzer} [ibid. 34, 207-241 (1981; Zbl 0426.90091)] is proposed. This policy is briefly compared with the principle of estimation and control recently obtained by \textit{M. Schäl} [Lect. Notes Pure Appl. Math. 86, 239-253 (1983; Zbl 0525.93071)].