Abstract
We investigate the convergence of finite-difference local descent algorithms for the solution of global optimization problems with a multi-extremum objective function. Application of noise-tolerant local descent algorithms to the class of so-called γ-regular problems makes it possible to bypass minor extrema and thus identify the global structure of the objective function. Experimental data presented in the article confirm the efficiency of the parallel gradient and coordinate descent algorithms for the solution of some test problems.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
REFERENCES
J. E. Dennis, Jr. and V. Torczon, “Direct search methods on parallel machines,” SIAM J. Optimization, 1, No. 4, 448-474 (1991).
G. Xue, “Molecular conformation on the CM-5 by parallel two-level simulated annealing,” J. Global Optimization, 4, 187-208 (1994).
A. Torn and S. Viitanen, “Topographical global optimization using pre-sampled points,” J. Global Optimization, 5, 267-276 (1994).
D. Bertsekas and Tsitsiklis, Parallel and Distributed Computation: Numerical Methods, Prentice Hall, Englewood Cliffs, NJ.
S. K. Zavriev, “On the global optimization properties of finite-difference local descent algorithms,” J. Global Optimization, 3, 63-78 (1993).
S. K. Zavriev and Yu. N. Perunova, “Using finite differences in the multistart method to solve a certain class of global optimization problems,” Vestnik MGU, Ser. 15, No. 3 (1997).
A. H. Rinnooy Khan and G. T. Timmer, “Stochastic global optimization methods,” Math. Programming, 39, 27-78 (1987).
A. Torn and A. Zilinskas, Global Optimization, Springer, Berlin (1989).
B. T. Polak, Introduction to Optimization [in Russian], Nauka, Moscow (1983).
Yu. N. Perunova, “Noise-tolerant methods for difference approximation of the gradient during parallelization,” electronic journal Issledovano v Rossii, 156, 1782-1791 (2001) http://zhurnal.ape.relarn.ru/articles/2001/156.pdf.
Y. Yong, “Dynamic tunneling algorithm for global optimization,” IEEE Trans. Systems, Man, Cybernetics, 19, No. 5, 1222-1230 (1989).
J. A. Northby, “Structure and binding of Lennard—Jones clusters: 13 ? N ? 147,” J. Chem. Phys., 87, No. 10, 6166-6177 (Nov. 1987).
T. C. Hu, V. Klee, and D. Larman, “Optimization of globally convex functions,” SIAM J. Control Optimization, 27, No. 5, 1026-1047 (1989).
D. Q. Mayne and E. Polak, “Nondifferentiable optimization via adaptive smoothing,” JOTA, 43, No. 4, 601-614 (1984).
C. D. Maranas and C. A. Floudas, “A deterministic global approach for molecular structure determination,” J. Chem. Phys., 100, No. 2, 1247-1261 (Jan. 1994).
R. Byrd, E. Eskow, and R. Schnabel, “A new large-scale global optimization method and its application to Lennard—Jones problems,” Tech. Rep. CU-CS-630-92, Dept. Computer Sci., Univ. of Colorado at Boulder.
A. Neumaier, “Molecular modeling of proteins and mathematical prediction of protein structure,” SIAM Rev., 39, No. 3, 407-460 (1997).
F. P. Vasil'ev, Numerical Methods for Solving Extremum Problems [in Russian], Nauka, Moscow (1980).
M. Hoare, “Structure and dynamics of simple microclusters,” Adv. Chem. Phys., 40, 49-135 (1979).
Rights and permissions
About this article
Cite this article
Zavriev, S.K., Perunova, Y.N. Parallel Versions of the Modified Coordinate and Gradient Descent Methods and Their Application to a Class of Global Optimization Problems. Computational Mathematics and Modeling 14, 108–122 (2003). https://doi.org/10.1023/A:1022951305895
Issue Date:
DOI: https://doi.org/10.1023/A:1022951305895