A modified nonlinear conjugate gradient algorithm for large-scale nonsmooth convex optimization (Q1985287)

Nonlinear conjugate gradient methods (CG) are among the most preferred for solving large-scale smooth problems because of their simplicity and low memory requirement. This paper presents the modified CG method, which includes the advantages of the CG methods for solving large-scale nonsmooth convex optimization problems. This method has the strong global convergence property of Dai and Yuan (DY) method and numerical efficiency of Hestenes and Stiefel (HS) method. The new method uses the search direction, which generates sufficient descent property and belongs to a trust region. Numerical results indicate that the method performs very well.