Generalized differentiation and duality in infinite dimensions under polyhedral convexity (Q2677672)

The authors investigate the theory of polyhedral duality in locally convex topological vector (LCTV) spaces and its applications. First, an LCTV generalization of the classical Rockafellar's proper separation theorem for two convex sets is derived. This is achieved by replacing the relative interior by the quasi-relative interior analogue. This result, along with the adoption of a geometric approach, is employed to derive enhanced calculus rules for normals to convex sets, coderivatives of convex set-valued mappings, and subgradients of extended-real-valued functions under certain polyhedrality requirements, in the setting of LCTV spaces. Further, new results on conjugate calculus and duality in convex optimization with relaxed qualification conditions in polyhedral settings, are obtained.