English

On the asymptotic convergence and acceleration of gradient methods

Optimization and Control 2019-08-21 v1

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.

Keywords

Cite

@article{arxiv.1908.07111,
  title  = {On the asymptotic convergence and acceleration of gradient methods},
  author = {Yakui Huang and Yu-Hong Dai and Xin-Wei Liu and Hongchao Zhang},
  journal= {arXiv preprint arXiv:1908.07111},
  year   = {2019}
}

Comments

28 pages, 7 figures

R2 v1 2026-06-23T10:51:39.127Z