arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1510.08573math.OC
收藏

关于优化梯度法的收敛性分析

On the convergence analysis of the optimized gradient method

Donghwan Kim, Jeffrey A. Fessler

摘要:

本文研究了具有已知Lipschitz常数的Lipschitz连续梯度的光滑凸函数的无约束极小化问题。我们最近提出了一个优化的梯度法(OGM),这个问题,并表明,它有一个最坏情况下的收敛界的费用函数下降,是Nesterov的快速梯度法(FGM)的两倍小,但有一个类似的有效的实际执行。Drori最近表明,OGM具有最佳的复杂性超过一般类的一阶方法。这种最优性使得充分研究OGM的收敛性变得非常重要。以前的最坏情况下的收敛界OGM推导出只有最后一个二次序列。本文给出了OGM生成的准素序列的一个解析收敛界。然后,我们讨论了OGM的其他收敛性质,包括有趣的事实,OGM有两种类型的最坏情况下的功能:分段仿射二次函数和二次函数。这些结果有助于完善光滑凸极小化的最优一阶方法理论。

英文摘要:

This paper considers the problem of unconstrained minimization of smooth convex functions having Lipschitz continuous gradients with known Lipschitz constant. We recently proposed an optimized gradient method (OGM) for this problem and showed that it has a worst-case convergence bound for the cost function decrease that is twice as small as that of Nesterov's fast gradient method (FGM), yet has a similarly efficient practical implementation. Drori showed recently that OGM has optimal complexity over the general class of first-order methods. This optimality makes it important to study fully the convergence properties of OGM. The previous worst-case convergence bound for OGM was derived for only the last iterate of a secondary sequence. This paper provides an analytic convergence bound for the primary sequence generated by OGM. We then discuss additional convergence properties of OGM, including the interesting fact that OGM has two types of worst-case functions: a piecewise affine-quadratic function and a quadratic function. These results help complete the theory of optimal first-order methods for smooth convex minimization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑