Convergence property of the Fletcher-Reeves conjugate gradient method with errors
C.Y. Wang, Mingkai Li · Journal of Industrial and Management Optimization · 2005
In this paper, we consider a new kind of Fletcher-Reeves (abbr. FR) conjugate gradient method witherrors, which is broadly applied in neural network training. Itsiterate formula is $x_{k+1}=x_{k}+\alpha_{k}(s_{k}+\omega_{k})$,where the main direction $s_{k}$ is obtained by FR conjugategradient method and $\omega_{k}$ is accumulative error. The globalconvergence property of the method is proved under the mildassumption conditions.