MULTIPLIER METHOD FOR PROXIMAL CONVEX OPTIMIZATION SEPARABLE

Abstract

The aim of this work is to prove the convergence of a proximal multiplicator method using generalized distances to solve convex minimization problems with separable structure, motivated in particular by the solution of optimization problems that arising in telecommunication networks and management of electrical energy production. The used procedures were the collection of information in scientific journals and specialized books, the study of the same and finally the use of mathematical tools to study the convergence of the sequence of the proposed method. The results show that, under some appropriate assumptions, the iterations generated by the method are well defined and the sequence converges to an optimal solution of the problem. Due to the generality of the study some papers related to proximal methods such as the works of Chen and Teboulle (1994), Kyono and Fukushima (2000) and Auslender and Teboulle (2001) are particular cases of our approach.

Keywords

• Métodos proximales
• problemas convexos separables
• ortante no negativo
• distancia generalizada
• funciones convexas.