Abstract
A general Markov model for discrete time adaptive control problems is considered, with compact state and action spaces. The unknown parameter is modeled as a random variable taking values also on a compact space. Under continuity assumptions on the transition operator of the Markov process, a nearly self optimizing control strategy is given. Two versions of such strategy are studied, with forced and randomized controls respectively
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have