Optimal solution control for a first-order partial stochastic differential equation (Q1280959): Difference between revisions

A linear first-order stochastic PDE is considered. The problem is to find an optimal control which minimizes some functional. The authors use a nice representation for the solution of the SPDE via an implicit function from the paper by \textit{Il. I. Gikhman} and \textit{T. M. Mestechkina} [Teor. Sluchajnykh Protsessov 11, 25-28 (1983; Zbl 0539.60054)]. It provides a covariance function of the solution. A measure density is used in order to get a lower bound for the cost functional. The optimal control is found explicitly in terms of a solution to a second order deterministic PDE. The reference to the density result is an unpublished preprint of \textit{G. A. Sokhadze} and \textit{A. D. Shatashvili} (1978); the result itself is some version of a Girsanov type formula for Gaussian fields. Remarks: I wonder if there's no mistake in the paper and whether it's improvable; perhaps some changes in the calculus could improve the exposition. On the other hand, it would take a lot of time to ``find all ends''.