Artículo de revista
Dual convergence for penalty proximal point algorithms in convex programming
Autor
Álvarez Daziano, Felipe
Carrasco, Miguel
Champion, Thierry
Institución
Resumen
We consider an implicit iterative method in convex programming which combines inexact
variants of the proximal point algorithm, with parametric penalty functions. We investigate a
multiplier sequence which is explicitly computed in terms of the primal sequence generated by the
iterative method, providing some conditions on the parameters in order to ensure convergence
towards a particular dual optimal solution.