Lagrange duality on DC evenly convex optimization problems via a generalized conjugation scheme

Abstract In this paper we study how Lagrange duality is connected to optimization problems whose objective function is the difference of two convex functions, briefly called DC problems. We present two Lagrange dual problems, each of them obtained via a different approach. While one of the duals corresponds to the standard formulation of the Lagrange dual problem, the other is written in terms of conjugate functions. When one of the involved functions in the objective is evenly convex, both problems are equivalent, but this relation is no longer true in the general setting. For this reason, we study conditions ensuring not only weak, but also zero duality gap and strong duality between the primal and one of the dual problems written using conjugate functions. For the other dual, and due to the fact that weak duality holds by construction, we just develop conditions for zero duality gap and strong duality between the primal DC problem and its (standard) Lagrange dual problem. Finally, we characterize weak and strong duality together with zero duality gap between the primal problem and its Fenchel-Lagrange dual following techniques used throughout the manuscript.