Some iterative methods of mathematical programming use a damping sequence {αt} such that 0 ≤ αt ≤ 1 for all t, αt → 0 as t → ∞, and Σ αt = ∞. For example, αt = 1/(t+1) in Brown's method for solving matrix games. In this paper, for a model class of iterative methods, the convergence rate for any damping sequence {αt} depending only on time t is computed. The computation is used to find the best damping sequence.
@article{urn:eudml:doc:41060,
title = {On the best choice of a damping sequence in iterative optimization methods.},
journal = {Publicacions Matem\`atiques},
volume = {32},
year = {1988},
pages = {275-287},
zbl = {0657.90100},
mrnumber = {MR0975902},
language = {en},
url = {http://dml.mathdoc.fr/item/urn:eudml:doc:41060}
}
Vaserstein, Leonid N. On the best choice of a damping sequence in iterative optimization methods.. Publicacions Matemàtiques, Tome 32 (1988) pp. 275-287. http://gdmltest.u-ga.fr/item/urn:eudml:doc:41060/