Methods of Conjugate Directions Versus Quasi-Newton Methods.
RESEARCH ANALYSIS CORP MCLEAN VA
Pagination or Media Count:
It is shown that algorithms for minimizing an unconstrained function Fx, x belongs to E sup n, which are solely methods of conjugate directions, can be expected to exhibit only an n or n - 1 step superlinear rate of convergence to an isolated local minimizer. This is contrasted with quasi-Newton methods which can be expected to exhibit every step superlinear convergence. Similar statements about a quadratic rate of convergence hold when a Lipschitz condition is placed on the second derivative of Fx. Author
- Theoretical Mathematics
- Operations Research