On Nonsmooth Problems of Stochastic Systems Optimization
Abstract
A class of stochastic optimization problems is analyzed that cannot be solved by deterministic and standard stochastic approximation methods. We consider risk control problems, optimization of stochastic networks and discrete event systems, screening irreversible changes, pollution control. The results of Ermoliev, Norkin, Wets [11] are extended to the case of problems involving random variables and general constraints. It is shown that the concept of mollifier subgradient leads to easily implementable computational procedures for stochastic systems with Lipschitz and discontinuous expectation functions. New optimality conditions are formulated enabling to design stochastic search procedures for constrained optimization of discontinuous systems.