A Modified Liu and Storey Conjugate Gradient Method for Large Scale Unconstrained Optimization Problems

The conjugate gradient method is one of the most popular methods to solve large-scale unconstrained optimization problems since it does not require second derivative, such as Newton’s or approximations. Moreover, can be applied in many fields neural networks, image restoration, etc. Many complicated are proposed these functions two three terms. In this paper, we propose a simple, easy, efficient, and robust method. new constructed based on Liu Storey overcome convergence problem descent property. modified satisfies properties sufficient condition under some assumptions. numerical results show that outperforms famous CG CG-Descent 5.3, Storey, Dai Liao. include number iterations CPU time.