Randomized Stochastic Approximation under Unknown but Bounded External Noise and It's Application to a Control of Educational Processes∗

In this paper the randomized stochastic approximation (RSA) methods for optimization problems for system unknown parameters tracking are studied. Such randomized SA procedures are working under arbitrary external observation noise. One of the possible application of the method to a control of educational processes is presented. Keywords — randomized stochastic approximation, simultaneous perturbation, unknown but bounded external noise, control of educational processes.