On the Convergence of the Entropy-Exponential Penalty Trajectories and Generalized Proximal Point Methods in Semidefinite Optimization

O. P. Ferreira (orizonmat.ufg.br)
P. R. Oliveira (poliveircos.ufrj.br)
R. C. M. Silva (rmesquitaufam.edu.br)

Abstract: The convergence of primal and dual central paths associated to entropy and exponential functions, respectively, for semidefinite programming problem are studied in this paper. As an application, the proximal point method with the Kullback-Leibler distance applied to semidefinite programming problems is considered, and the convergence of primal and dual sequences is proved.

Keywords: generalized proximal point methods, Bregman distances, central path, semidefinite programming.

Category 1: Linear, Cone and Semidefinite Programming

Category 2: Linear, Cone and Semidefinite Programming (Semi-definite Programming )

Citation:

Download: [PDF]

Entry Submitted: 07/10/2006
Entry Accepted: 07/10/2006
Entry Last Modified: 06/18/2007

Modify/Update this entry