Abstract
The paper is devoted to the convergence properties of finite-difference local descent algorithms in global optimization problems with a special γ-convex structure. It is assumed that the objective function can be closely approximated by some smooth convex function. Stability properties of the perturbed gradient descent and coordinate descent methods are investigated. Basing on this results some global optimization properties of finite-difference local descent algorithms, in particular, coordinate descent method, are discovered. These properties are not inherent in methods using exact gradients.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Hu, T. C., V. Klee and D. Larman (1989), Optimization of Globally Convex Functions,SIAM J. Control and Optimization 27 (5), 1026–1047.
Mayne, D. Q. and E. Polak (1984), Nondifferential Optimization via Adaptive Smoothing,JOTA 43 (4), 601–614.
Mikhalevitch, V. S., A. M. Gupal and V. I. Norkin (1987),Methods of Nonconvex Optimization, Nauka, Moscow (in Russian).
Boggs, P. T. and J. E. Dennis, Jr. (1976), A Stability Analysis for Perturbed Nonlinear Iterative Methods,Mathematics of Computation 30 (134), 199–215.
Polyak, B. T. (1987),Introduction to Optimization. Optimization Softwear, Inc., NY.
Zavriev, S. K. (1984),Stochastic Sub-Gradients Methods for Minmax Problems. Izd. MGU, Moscow, (1984) (in Russian).
Dorofeev, P. A. (1985), On Some Properties of Quasigradient Method,USSR Computational Mathematics and Mathematical Physics 25, 181–189.
Zavriev, S. K. (1990), Convergence Properties of the Gradient Method in Presence of Perturbations of a Variable Level,USSR Computational Mathematics and Mathematical Physics 30, 997–1007.
Norkin, V. I. (1979), On the Convergence Conditions for Combined Nonlinear Programming Algorithms,Kibernetica 6, 73–77.
Luenberger, D. G. (1984),Linear and Nonlinear Programming, Addison-Wesley, London.
Evtushenko, U. G. (1971), Numerical Method for the Search of Global Extremum of Function,USSR Computational Mathematics and Mathematical Physics 11, 1390–1403.
Horst, R. (1986), A General Class of Branch-and-Bound Methods in Global Optimization with Some New Approaches for Concave Minimization,Journal of Optimization Theory and Applications 51, 271–291.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Zavriev, S.K. On the global optimization properties of finite-difference local descent algorithms. J Glob Optim 3, 67–78 (1993). https://doi.org/10.1007/BF01100240
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF01100240