PROJECT TITLE :

Distributed Stochastic Optimization via Correlated Scheduling

ABSTRACT:

This paper considers a drawback where multiple devices make repeated decisions primarily based on their own observed events. The events and choices at every time-step determine the values of a utility perform and a collection of penalty functions. The goal is to make distributed choices over time to maximize time-average utility subject to time-average constraints on the penalties. An example may be a assortment of power-constrained sensors that repeatedly report their own observations to a fusion center. Most time-average utility is basically reduced as a result of devices don't know the events observed by others. Optimality is characterised for this distributed context. It is shown that optimality is achieved by correlating device selections through a commonly known pseudo-random sequence. An optimal algorithm is developed that chooses pure ways at each time-step based mostly on a set of time-varying weights.