On nondifferentiable second order symmetric duality in mathematical programming (Q1766383)

This article discusses issues of duality in non-differentiable, nonlinear mathematical programs. In nonlinear systems, the dual of the dual is not necessarily the primal problem, and when this property is true then the formulation is called symmetric. The authors study in detail non-differentiable second order symmetric duality under the Wolfe type \(\eta\)-bonvexity/boncavity. The paper begins with a series of Theorems and useful propositions that are proven in full. In section 4 duality in minimax mixed integer programming models is discussed. The paper concludes with a note on self duality (when a mathematical program is identical to its dual), and a small number of references.