Many of the traditional optimization algorithms have their direct counterparts in Riemannian space. This paper concentrates on gradient based algorithms, in particular the generalized versions of gradient ascent and conjugate gradient method.
