Geodesic
Gradient method with dynamical retards for large-scale optimization problems
Electronic transactions on numerical analysis, Tome 16 (2003), pp. 186-193
We consider a generalization of the gradient method with retards for the solution of large-scale unconstrained optimization problems. Recently, the gradient method with retards was introduced to find global minimizers of large-scale quadratic functions. The most interesting feature of this method is that it does not involve a decrease in the objective function, which allows fast local convergence. On the other hand, nonmonotone globalization strategies, that preserve local behavior for the nonquadratic case, have proved to be very effective when associated with low storage methods. In this work, the gradient method with retards is generalized and combined in a dynamical way with nonmonotone globalization strategies to obtain a new method for minimizing nonquadratic functions, that can deal efficiently with large problems. Encouraging numerical experiments on well-known test problems are presented.
Classification : 49M07, 49M10, 65K
Keywords: spectral gradient method, nonmonotone line search, barzilai-borwein method, polak-ribi`ere method, Rayleigh quotient 
@article{ETNA_2003__16__a0,
     author = {Luengo,  Francisco and Raydan,  Marcos},
     title = {Gradient method with dynamical retards for large-scale optimization problems},
     journal = {Electronic transactions on numerical analysis},
     pages = {186--193},
     year = {2003},
     volume = {16},
     zbl = {1134.90496},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/ETNA_2003__16__a0/}
}
TY  - JOUR
AU  - Luengo,  Francisco
AU  - Raydan,  Marcos
TI  - Gradient method with dynamical retards for large-scale optimization problems
JO  - Electronic transactions on numerical analysis
PY  - 2003
SP  - 186
EP  - 193
VL  - 16
UR  - http://geodesic.mathdoc.fr/item/ETNA_2003__16__a0/
LA  - en
ID  - ETNA_2003__16__a0
ER  - 
%0 Journal Article
%A Luengo,  Francisco
%A Raydan,  Marcos
%T Gradient method with dynamical retards for large-scale optimization problems
%J Electronic transactions on numerical analysis
%D 2003
%P 186-193
%V 16
%U http://geodesic.mathdoc.fr/item/ETNA_2003__16__a0/
%G en
%F ETNA_2003__16__a0
Luengo,  Francisco; Raydan,  Marcos. Gradient method with dynamical retards for large-scale optimization problems. Electronic transactions on numerical analysis, Tome 16 (2003), pp. 186-193. http://geodesic.mathdoc.fr/item/ETNA_2003__16__a0/