Previous |  Up |  Next

Article

Title: A new diagonal quasi-Newton algorithm for unconstrained optimization problems (English)
Author: Nosrati, Mahsa
Author: Amini, Keyvan
Language: English
Journal: Applications of Mathematics
ISSN: 0862-7940 (print)
ISSN: 1572-9109 (online)
Volume: 69
Issue: 4
Year: 2024
Pages: 501-512
Summary lang: English
.
Category: math
.
Summary: We present a new diagonal quasi-Newton method for solving unconstrained optimization problems based on the weak secant equation. To control the diagonal elements, the new method uses new criteria to generate the Hessian approximation. We establish the global convergence of the proposed method with the Armijo line search. Numerical results on a collection of standard test problems demonstrate the superiority of the proposed method over several existing diagonal methods. (English)
Keyword: unconstrained optimization
Keyword: diagonal quasi-Newton method
Keyword: weak secant equation
Keyword: global convergence
MSC: 65K05
MSC: 90C30
DOI: 10.21136/AM.2024.0045-24
.
Date available: 2024-08-27T11:19:13Z
Last updated: 2024-09-02
Stable URL: http://hdl.handle.net/10338.dmlcz/152531
.
Reference: [1] Andrei, N.: An acceleration of gradient descent algorithm with backtracking for unconstrained optimization.Numer. Algorithms 42 (2006), 63-73. Zbl 1101.65058, MR 2249567, 10.1007/s11075-006-9023-9
Reference: [2] Andrei, N.: An unconstrained optimization test functions collection.Adv. Model. Optim. 10 (2008), 147-161. Zbl 1161.90486, MR 2424936
Reference: [3] Andrei, N.: A diagonal quasi-Newton updating method based on minimizing the measure function of Byrd and Nocedal for unconstrained optimization.Optimization 67 (2018), 1553-1568. Zbl 1402.65049, MR 3877965, 10.1080/02331934.2018.1482298
Reference: [4] Andrei, N.: A diagonal quasi-Newton updating method for unconstrained optimization.Numer. Algorithms 81 (2019), 575-590. Zbl 1416.49025, MR 3953161, 10.1007/s11075-018-0562-7
Reference: [5] Andrei, N.: A new accelerated diagonal quasi-Newton updating method with scaled forward finite differences directional derivative for unconstrained optimization.Optimization 70 (2021), 345-360. Zbl 1460.90204, MR 4207210, 10.1080/02331934.2020.1712391
Reference: [6] Armijo, L.: Minimization of functions having Lipschitz continuous first partial derivatives.Pac. J. Math. 16 (1966), 1-3. Zbl 0202.46105, MR 0191071, 10.2140/pjm.1966.16.1
Reference: [7] Barzilai, J., Borwein, J. M.: Two-point step size gradient methods.IMA J. Numer. Anal. 8 (1988), 141-148. Zbl 0638.65055, MR 0967848, 10.1093/imanum/8.1.141
Reference: [8] Bongartz, I., Conn, A. R., Gould, N., Toint, P. L.: CUTE: Constrained and unconstrained testing environment.ACM Trans. Math. Softw. 21 (1995), 123-160. Zbl 0886.65058, 10.1145/200979.201043
Reference: [9] C. G. Broyden, J. E. Dennis, Jr., J. J. Moré: On the local and superlinear convergence of quasi-Newton methods.J. Inst. Math. Appl. 12 (1973), 223-245 \99999DOI99999 10.1093/imamat/12.3.223 . Zbl 0282.65041, MR 0341853, 10.1093/imamat/12.3.223
Reference: [10] J. E. Dennis, Jr., J. J. Moré: A characterization of superlinear convergence and its application to quasi-Newton methods.Math. Comput. 28 (1974), 549-560. Zbl 0282.65042, MR 0343581, 10.1090/S0025-5718-1974-0343581-1
Reference: [11] J. E. Dennis, Jr., J. J. Moré: Quasi-Newton methods, motivation and theory.SIAM Rev. 19 (1977), 46-89. Zbl 0356.65041, MR 0445812, 10.1137/101900
Reference: [12] J. E. Dennis, Jr., H. Wolkowicz: Sizing and least-change secant methods.SIAM J. Numer. Anal. 30 (1993), 1291-1314. Zbl 0802.65081, MR 1239822, 10.1137/073006
Reference: [13] Dolan, E. D., Moré, J. J.: Benchmarking optimization software with performance profiles.Math. Program. 91 (2002), 201-213. Zbl 1049.90004, MR 1875515, 10.1007/s101070100263
Reference: [14] Farid, M., Leong, W. J., Zheng, L.: A new diagonal gradient-type method for large scale unconstrained optimization.Sci. Bull., Ser. A, Appl. Math. Phys., Politeh. Univ. Buchar. 75 (2013), 57-64. Zbl 1299.65118, MR 3032542
Reference: [15] Gill, P. E., Murray, W.: Conjugate-gradient methods for large-scale nonlinear optimization.Technical Report SOL-79-15 Stanford University, Stanford (1979), 1-66.
Reference: [16] Goldstein, A. A.: On steepest descent.J. Soc. Ind. Appl. Math., Ser. A: Control 3 (1965), 147-151. Zbl 0221.65094, MR 0184777, 10.1137/030301
Reference: [17] Leong, W. J., Enshaei, S., Kek, S. L.: Diagonal quasi-Newton methods via least change updating principle with weighted Frobenius norm.Numer. Algorithms 86 (2021), 1225-1241. Zbl 1464.90099, MR 4211118, 10.1007/s11075-020-00930-9
Reference: [18] Leong, W. J., Farid, M., Hassan, M. A.: Improved Hessian approximation with modified quasi-Cauchy relation for a gradient-type method.Adv. Model. Optim. 12 (2010), 37-44. Zbl 1332.90346, MR 2591783
Reference: [19] Nash, S. G.: Preconditioning of truncated-Newton methods.SIAM J. Sci. Stat. Comput. 6 (1985), 599-616. Zbl 0592.65038, MR 0791188, 10.1137/0906042
Reference: [20] Nocedal, J.: Updating quasi-Newton matrices with limited storage.Math. Comput. 35 (1980), 773-782. Zbl 0464.65037, MR 0572855, 10.1090/S0025-5718-1980-0572855-7
Reference: [21] Nocedal, J., Wright, S. J.: Numerical Optimization.Springer Series in Operations Research and Financial Engineering. Springer, New York (2006). Zbl 1104.65059, MR 2244940, 10.1007/b98874
Reference: [22] Powell, M. J. D.: A new algorithm for unconstrained optimization.Nonlinear Programming Elsevier, Amsterdam (1970), 31-65 \99999DOI99999 10.1016/B978-0-12-597050-1.50006-3 . Zbl 0228.90043, MR 0272162
Reference: [23] Raydan, M.: The Barzilai and Borwein gradient method for the large scale unconstrained minimization problem.SIAM J. Optim. 7 (1997), 26-33. Zbl 0898.90119, MR 1430555, 10.1137/S1052623494266365
Reference: [24] Wolfe, P.: Convergence conditions for ascent methods.SIAM Rev. 11 (1969), 226-235. Zbl 0177.20603, MR 0250453, 10.1137/101103
Reference: [25] Zhu, M., Nazareth, J. L., Wolkowicz, H.: The quasi-Cauchy relation and diagonal updating.SIAM J. Optim. 9 (1999), 1192-1204. Zbl 1013.90137, MR 1724783, 10.1137/S1052623498331793
.

Fulltext not available (moving wall 24 months)

Partner of
EuDML logo