Optimal Markov monotone symmetric random search (Q1571313)

The paper is devoted to the construction of optimal Markov random search methods. Optimal in a broad class of optimization methods, these methods have a simple structure, i.e., these methods are described by a finite number of parameters. The results obtained imply that one can reduce the number of methods to consider and, at the same time, retain optimal properties. Serving as a criterion for the choice of a search method, the theoretical result presented is of great practical value.