Descent methods for convex optimization problems in Banach spaces.
2000 Mathematics Subject Classification: 90C48, 49N15, 90C25In this paper we reconsider a nonconvex duality theory established by B. Lemaire and M. Volle (see [4]), related to a primal problem of minimizing the difference of two convex functions subject to a DC-constraint. The purpose of this note is to present a new method based on Toland-Singer duality principle. Applications to the case when the constraints are vector-valued are provided.