scientific article; zbMATH DE number 3924660

From MaRDI portal
Publication:3699592





[For the entire collection see Zbl 0568.00025.] The paper surveys the recent investigations of the following procedure to control Markov processes with unknown parameters: the parameters are continuously estimated from the observed trajectory, the optimal stationary policy corresponding to the estimates is found, and the control parameter is adjusted to the value prescribed by that policy. The basic ideas are explained by the problem of preventive renewals. Then finite state Markov processes, Markov chains and diffusion processes are considered.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3699592)