Adaptive Policies for Discrete-Time Stochastic Control Systems with Unknown Disturbance Distribution

摘要

The authors introduce adaptive policies for discrete-time, infinite horizon, stochastic control systems with discounted reward criterion, where the disturbance process is a sequence of independent and identically distributed random elements with unknown distribution. These policies are shown to be asymptotically optimal and for each of them we obtain (almost surely) uniform approximations of the optimal reward function. Keywords: Reprints. (kr)