Sample-path solutions for simulation optimization problems and stochastic variational inequalities.

*(English)*Zbl 0893.90137
Woodruff, David L. (ed.), Advances in computational and stochastic optimization, logic programming, and heuristic search. Interfaces in computer science and operations research. 6th conference of the Computer Science Technical Section of INFORMS, January 7–9, 1998, Carmel, CA, USA. Boston: Kluwer Academic Publishers. Oper. Res./Comput. Sci. Interfaces Ser. 9, 169-188 (1998).

Summary: We give an overview of some recent developments in using simulation together with gradient estimation techniques to provide solutions for difficult stochastic optimization problems and stochastic variational inequalities. The basic idea is to observe a fixed sample path (by using the method of common random numbers from the simulation literature), solve the resulting deterministic problem using fast and effective methods from nonlinear programming, and then use the resulting solutions to infer information about the solution of the original stochastic problem. We describe these so-called sample-path methods precisely, review some conditions under which they are known to work, and comment on their potential advantages and limitations. We also illustrate some application areas in which these ideas have been successful.

For the entire collection see [Zbl 0881.00042].

For the entire collection see [Zbl 0881.00042].

##### Software:

PATH Solver
PDF
BibTeX
XML
Cite

\textit{G. Gürkan} et al., in: Advances in computational and stochastic optimization, logic programming, and heuristic search. Interfaces in computer science and operations research. 6th conference of the Computer Science Technical Section of INFORMS, January 7--9, 1998, Carmel, CA, USA. Boston: Kluwer Academic Publishers. 169--188 (1998; Zbl 0893.90137)