Abstract
General convergence characteristics of stochastic optimization methods are investigated, when convex and/or differentiable structure of the optimization problems to be solved is not assumed, First, some basic stochastic optimization schemes are introduced and their convergence properties are analysed, then the obtained results are extended for the case of stochastically combined (hybrid) procedures. Finally, some experimental results with hybrid optimization methods are summarized.

This publication has 8 references indexed in Scilit: