Mathematical Conferences Niš, Serbia, 13th Serbian Mathematical Congress

Font Size: 
New Hybrid Conjugate Gradient Method for Unconstrained Optimization
Snezana Djordjevic

Last modified: 2014-02-05

Abstract


A new hybrid conjugate gradient algorithm is considered. Werequire a specific determination of parameters. The parameter$\beta_k$ is computed as a convex combination of $\beta_k^{LS}$(Liu and Storey) and $\beta_k^{CD}$ (Conjugate Descent). Wecompute the parameter $\theta_k$  in such a way that the conjugacycondition is satisfied. The standard Wolfe line search conditionsare used. Numerical comparisons show that the  present hybridconjugate gradient algorithm often behaves better than someprevious algorithms.