Abstract
In this paper, we propose an improved trust region method for solving unconstrained optimization problems. Different with traditional trust region methods, our algorithm does not resolve the subproblem within the trust region centered at the current iteration point, but within an improved one centered at some point located in the direction of the negative gradient, while the current iteration point is on the boundary set. We prove the global convergence properties of the new improved trust region algorithm and give the computational results which demonstrate the effectiveness of our algorithm.
Similar content being viewed by others
References
Conn A R, Gould N I M, Toint Ph L. Trust-Region Methods. Philadelphia: SIAM, 2000
Corradi G. Quasi-Newton methods for nonlinear equations and unconstrained optimization methods. Intern J Comput Math, 1991, 38: 71–89
Fan J Y, Yuan Y X. A new trust region algorithm with trust region radius converging to zero. In: Proceedings of the 5th International Conference on Optimization: Techniques and Applications. Hong Kong: Contemporary Development Company, 2001, 786–794
Fletcher R. Practical Methods of Optimization. New York: John Wiley and Sons, 1987
Gay D M. Computing optimal locally constrained steps. SIAM J Sci Statist Comput, 1981, 2: 186–197
Gill P E, Murray W. Quasi-Newton methods for unconstrained optimization. J Inst Maths Appl, 1972, 9: 91–108
Hei L. A self-adaptive trust region algorithm. J Comput Math, 2003, 21: 229–236
Khalfan H F H. Topics in quasi-Newton methods for unconstrained optimization. PhD Thesis. Colorado: University of Colorado, 1989
Li C J, Sun W Y. On filter-successive linearization methods for nonlinear semidefinite programming. Sci China Ser A, 2009, 52: 2341–2361
Ma G X. A modified trust region methods for unconstrained optimization. Master Thesis. Beijing: Academy of Mathematics and Systems Science, CAS, 2003
Morales J L. A trust region based algorithm for unconstrained derivative-free optimization. Technical Report, Departamento de Matemáticas ITAM, 2007
Moré J J, Garbow B S, Hillstrom K H. Testing unconstrained optimization software. ACM Trans Math Software, 1981, 7: 17–41
Moré J J, Sorensen D C. Computing a trust region step. SIAM J Sci Statist Comput, 1983, 4: 553–572
Nocedal J, Yuan Y. Combining trust region and line search. In: Yuan Y, eds. Advances in Nonlinear Programming. Dordrecht: Kluwer, 1998, 153–175
Powell M J D. Convergence properties of a class of minimization algorithms. In: Mangasarian O L, Meyer R R, Robinson S M, eds. Nonlinear Programming. New York: Academic Press, 1975, 1–27
Shultz G A, Schnabel R B, Byrd R H. A family of trust-region-based algorithms for unconstrained minimization with strong global convergence. SIAM J Numer Anal, 1985, 22: 47–67
Sun W Y, Zhou Q Y. An unconstrained optimization method using nonmonotone second order Goldstein’s line search. Sci China Ser A, 2007, 50: 1389–1400
Wang J Y, Ni Q. An algorithm for solving new trust region subproblem with conic model. Sci China Ser A, 2008, 51: 461–473
Yuan G L, Wei Z X. New line search methods for unconstrained optimization. J Korean Statist Soc, 2009, 38: 29–39
Yuan Y. A review of trust region algorithms for optimization. In: Ball J M, Hunt J C R, eds. ICM99: Proceedings of the Fourth International Congress on Industrial and Applied Mathematics. Oxford: Oxford University Press, 2000, 271–282
Zhang X S, Chen Z W, Zhang J L. A self-adaptive trust region method for unconstrained optimization. OR Transactions, 2001, 5: 53–62
Zhou Q H, Li Y, Ha M H. Constructing composite search directions with parameters in quadratic interpolation models. J Global Optim, 2011, 51: 313–324
Zhou Q H. On the use of simplex methods in constructing quadratic models. Sci China Ser A, 2007, 50: 913–924
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhou, Q., Zhang, Y., Xu, F. et al. An improved trust region method for unconstrained optimization. Sci. China Math. 56, 425–434 (2013). https://doi.org/10.1007/s11425-012-4507-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11425-012-4507-7
Keywords
- unconstrained optimization
- trust region methods
- global convergence
- negative gradient direction
- iterative