Optimality conditions of generalized subconvexlike set-valued optimization problems based on the quasi-relative interior (Q639941)

A usual assumption in mathematical programming is that the interior of a convex cone is nonempty. However, it can happen that the interior of a convex cone is empty. To overcome the difficulty, it is necessary to weaken the notion of the interior. In this paper, some properties of the quasi-relative interior are studied. Using the quasi-relative interior, the authors define a class of generalized convex set-valued maps named generalized subconvexlike set-valued maps. Some separation properties are obtained by a separation theorem involving the quasi-relative interior. Some optimality conditions are established, too. Comparing the quasi-relative interior with the other interiors, the authors find that the quasi-relative interior is a kind of very weak interior.