Penalization in non-classical convex programming via variational convergence (Q1181900)

This is a penalty method for approximating constrained optimization problems by unconstrained ones, in the case that more general convex functions are considered which cannot necessarily be extended outside their effective domains by an everywhere finite convex function. The concepts of the perturbed proximal regularization method and variational convergence are described, and applied to the problem, in conjunction with a new penalty method. Extensive computational results are given.