Optimization theory for n-set functions (Q752016)

The author develops an optimization theory for real-valued functions depending on n sets instead of n real variables. The results consist of differential conditions for local minima, saddle point optimality conditions and a Lagrangian duality. They are based on results of \textit{R. J. T. Morris} [J. Math. Anal. App. 70, No.2, 546-562 (1979; Zbl 0417.49032)], where functions of a single set are investigated, and on proof techniques as in the book by \textit{D. G. Luenberger} [``Optimization by vector space methods'' (1969; Zbl 0176.127)].