Quadratic Regularization for Global Optimization.

In this paper we present a novel global optimization method for solving continuous nonlinear optimization problems. This method is based on an exact quadratic regularization (EQR). It allows the given problems to convert to equivalent problem maximization the norm of a vector on a convex set. Such problems are easier to solve than general nonlinear optimization problems. For the solution of these problems we use only primal-dual interior point method and a dichotomy search. The comparative numerical experiments have proved that EQR method to be very efficient. [ABSTRACT FROM AUTHOR]