A new hybrid generalized proximal point algorithm for variational inequality problems (Q1422877)

The author considers the single-valued variational inequality in the finite-dimensional Euclidean space setting and describes a two-level iterative scheme where an inexact iteration of a proximal point method with Bregman functions is used for computing parameters of a separating hyperplane and for providing the the Fejer-monotone convergence to a solution. The general two-level scheme with various auxiliary procedures for determining the separating hyperplanes was suggested by \textit{I.V.~Konnov} [Russ. Math. 37, No. 2, 44--51 (1993; Zbl 0835.90123)]. The algorithm involves the approximate criterion suggested by \textit{M. V.~Solodov} and \textit{B. F. Svaiter} [Set-Valued Anal. 7, 323--345 (1999; Zbl 0959.90038)]. This criterion is shown to be satisfied by a single Newton-type step. Some results of computational experiments are also given.