On the Linear Convergence of Forward–Backward Splitting Method: Part I—Convergence Analysis

On the Linear Convergence of Forward–Backward Splitting Method: Part I—Convergence Analysis
复制标题

前向-后向分裂法的线性收敛性:第一部分-收敛性分析

DOI:
10.1007/s10957-020-01787-7
复制
发表时间:
2021
影响因子:
1.9
通讯作者:
Nghia, Tran T.
Nghia, Tran T.
中科院分区:
数学3区
文献类型:
--
作者:
Bello-Cruz, Yunier;Li, Guoyin;Nghia, Tran T.

文献摘要

参考文献

被引文献

相似文献

本文研究了求解凸优化问题的Beck-Teboulle线搜索前后向分裂方法的复杂性,其中目标函数可分解为一个可微函数和一个非光滑函数之和.我们表明,该方法弱收敛到Hilbert空间中的最优解,在温和的常设假设没有全球Lipschitz连续性的可微函数的梯度参与。我们的常设假设比Salzo论文中的相应条件弱(SIAM J Optim 27:2153-2181,2017)。在可微函数梯度局部Lipschitz连续的条件下,得到了函数值次线性收敛的传统复杂性。我们的主要结果是关于线性收敛的方法(商型),无论是在函数值序列和迭代序列,只有二次增长的条件。我们的证明技术是直接从二次增长的条件和一些性质的向前向后分裂方法,而不使用误差界或Kurdya-Eschojasiewicz不等式在其他出版物在这个方向。
In this paper, we study the complexity of the forward–backward splitting method with Beck–Teboulle’s line search for solving convex optimization problems, where the objective function can be split into the sum of a differentiable function and a nonsmooth function. We show that the method converges weakly to an optimal solution in Hilbert spaces, under mild standing assumptions without the global Lipschitz continuity of the gradient of the differentiable function involved. Our standing assumptions is weaker than the corresponding conditions in the paper of Salzo (SIAM J Optim 27:2153–2181, 2017). The conventional complexity of sublinear convergence for the functional value is also obtained under the local Lipschitz continuity of the gradient of the differentiable function. Our main results are about the linear convergence of this method (in the quotient type), in terms of both the function value sequence and the iterative sequence, under only the quadratic growth condition. Our proof technique is direct from the quadratic growth conditions and some properties of the forward–backward splitting method without using error bounds or Kurdya-Łojasiewicz inequality as in other publications in this direction.
通过度量的变化实现非线性局部误差范围
DOI: --
发表时间: 2014
期刊:
影响因子: --
作者:
D. Azé;J. Corvellec
通讯作者: J. Corvellec
DOI: 10.1287/moor.2017.0889
发表时间: 2016-02
期刊: Math. Oper. Res.
影响因子: --
作者:
D. Drusvyatskiy;A. Lewis
通讯作者: D. Drusvyatskiy;A. Lewis
希尔伯特空间中的阈值梯度方法:支持辨识和线性收敛
DOI: --
发表时间: 2017
期刊: E S A I M: Control, Optimisation and Calculus of Variations
影响因子: --
作者:
Guillaume Garrigos;L. Rosasco;S. Villa
通讯作者: S. Villa
近端前向后向分裂方法与线搜索的收敛性研究
DOI: --
发表时间: 2015
期刊:
影响因子: --
作者:
Yunier Bello Cruz;T. Nghia
通讯作者: T. Nghia