Abstract
We consider the asymptotic behavior of a family of gradient methods, which include the steepest descent and minimal gradient methods as special instances. It is proved that each method in the family will asymptotically zigzag between two directions. Asymptotic convergence results of the objective value, gradient norm, and stepsize are presented as well. To accelerate the family of gradient methods, we further exploit spectral properties of stepsizes to break the zigzagging pattern. In particular, a new stepsize is derived by imposing finite termination on minimizing two-dimensional strictly convex quadratic function. It is shown that, for the general quadratic function, the proposed stepsize asymptotically converges to the reciprocal of the largest eigenvalue of the Hessian. Furthermore, based on this spectral property, we propose a periodic gradient method by incorporating the Barzilai-Borwein method. Numerical comparisons with some recent successful gradient methods show that our new method is very promising.
Similar content being viewed by others
References
Akaike, H.: On a successive transformation of probability distribution and its application to the analysis of the optimum gradient method. Ann. Inst. Stat. Math. 11(1), 1–16 (1959)
Barzilai, J., Borwein, J.M.: Two-point step size gradient methods. IMA J. Numer. Anal. 8(1), 141–148 (1988)
Birgin, E.G., Martínez, J.M., Raydan, M.: Nonmonotone spectral projected gradient methods on convex sets. SIAM J. Optim. 10(4), 1196–1211 (2000)
Cauchy, A.: Méthode générale pour la résolution des systemes déquations simultanées. Comp. Rend. Sci. Paris 25, 536–538 (1847)
Dai, Y.H.: Alternate step gradient method. Optimization 52(4–5), 395–415 (2003)
Dai, Y.H., Fletcher, R.: On the asymptotic behaviour of some new gradient methods. Math. Program. 103(3), 541–559 (2005)
Dai, Y.H., Huang, Y., Liu, X.W.: A family of spectral gradient methods for optimization. Comput. Optim. Appl. 74(1), 43–65 (2019)
Dai, Y.H., Liao, L.Z.: \(R\)-linear convergence of the Barzilai and Borwein gradient method. IMA J. Numer. Anal. 22(1), 1–10 (2002)
Dai, Y.H., Yang, X.: A new gradient method with an optimal stepsize property. Comput. Optim. Appl. 33(1), 73–88 (2006)
Dai, Y.H., Yuan, Y.X.: Alternate minimization gradient method. IMA J. Numer. Anal. 23(3), 377–393 (2003)
Dai, Y.H., Yuan, Y.X.: Analysis of monotone gradient methods. J. Ind. Mang. Optim. 1(2), 181 (2005)
De Asmundis, R., Di Serafino, D., Hager, W.W., Toraldo, G., Zhang, H.: An efficient gradient method using the Yuan steplength. Comput. Optim. Appl. 59(3), 541–563 (2014)
De Asmundis, R., di Serafino, D., Riccio, F., Toraldo, G.: On spectral properties of steepest descent methods. IMA J. Numer. Anal. 33(4), 1416–1435 (2013)
Di Serafino, D., Ruggiero, V., Toraldo, G., Zanni, L.: On the steplength selection in gradient methods for unconstrained optimization. Appl. Math. Comput. 318, 176–195 (2018)
Dolan, E.D., Moré, J.J.: Benchmarking optimization software with performance profiles. Math. Program. 91(2), 201–213 (2002)
Elman, H.C., Golub, G.H.: Inexact and preconditioned Uzawa algorithms for saddle point problems. SIAM J. Numer. Anal. 31(6), 1645–1661 (1994)
Fletcher, R.: On the Barzilai-Borwein method. In: Optimization and Control with Applications, pp. 235–256. Springer, Boston (2005)
Forsythe, G.E.: On the asymptotic directions of the s-dimensional optimum gradient method. Numer. Math. 11(1), 57–76 (1968)
Frassoldati, G., Zanni, L., Zanghirati, G.: New adaptive stepsize selections in gradient methods. J. Ind. Mang. Optim. 4(2), 299 (2008)
Gonzaga, C.C., Schneider, R.M.: On the steepest descent algorithm for quadratic functions. Comput. Optim. Appl. 63(2), 523–542 (2016)
Huang, Y., Dai, Y.H., Liu, X.W., Zhang, H.: Gradient methods exploiting spectral properties. Optim. Method Softw. 35(4), 681–705 (2020)
Huang, Y., Liu, H.: Smoothing projected Barzilai-Borwein method for constrained non-Lipschitz optimization. Comput. Optim. Appl. 65(3), 671–698 (2016)
Huang, Y., Liu, H., Zhou, S.: Quadratic regularization projected Barzilai-Borwein method for nonnegative matrix factorization. Data Min. Knowl. Disc. 29(6), 1665–1684 (2015)
Jiang, B., Dai, Y.H.: Feasible Barzilai-Borwein-like methods for extreme symmetric eigenvalue problems. Optim. Method Softw. 28(4), 756–784 (2013)
Liu, Y.F., Dai, Y.H., Luo, Z.Q.: Coordinated beamforming for miso interference channel: Complexity analysis and efficient algorithms. IEEE Trans. Signal Process. 59(3), 1142–1157 (2011)
Nocedal, J., Sartenaer, A., Zhu, C.: On the behavior of the gradient norm in the steepest descent method. Comp. Optim. Appl. 22(1), 5–35 (2002)
Pronzato, L., Wynn, H.P., Zhigljavsky, A.A.: Asymptotic behaviour of a family of gradient algorithms in \(R^d\) and Hilbert spaces. Math. program. 107(3), 409–438 (2006)
Raydan, M.: On the Barzilai and Borwein choice of steplength for the gradient method. IMA J. Numer. Anal. 13(3), 321–326 (1993)
Raydan, M.: The Barzilai and Borwein gradient method for the large scale unconstrained minimization problem. SIAM J. Optim. 7(1), 26–33 (1997)
Sun, C., Liu, J.P.: New stepsizes for the gradient method. Optim. Lett. 14, 1943–1955 (2020)
Yuan, Y.X.: A new stepsize for the steepest descent method. J. Comput. Math. 24(2), 149–156 (2006)
Yuan, Y.X.: Step-sizes for the gradient method. AMS IP Stud. Adv. Math. 42(2), 785–796 (2008)
Zhou, B., Gao, L., Dai, Y.H.: Gradient methods with adaptive step-sizes. Comput. Optim. Appl. 35(1), 69–86 (2006)
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grant Nos. 11701137, 11631013, 12071108, 11671116, 11991021, 12021001), the Strategic Priority Research Program of Chinese Academy of Sciences (Grant No. XDA27000000), Beijing Academy of Artificial Intelligence (BAAI), Natural Science Foundation of Hebei Province (Grant No. A2021202010), China Scholarship Council (Grant No. 201806705007), and USA National Science Foundation (Grant Nos. DMS-1819161, DMS-2110722).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Huang, Y., Dai, YH., Liu, XW. et al. On the Asymptotic Convergence and Acceleration of Gradient Methods. J Sci Comput 90, 7 (2022). https://doi.org/10.1007/s10915-021-01685-8
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10915-021-01685-8