用于最小化可微函数与凸函数之和的辅助问题原理与非精确变度量前向-后向算法
最优化与控制
2016-09-13 v2
摘要
针对最小化可微函数与凸函数之和的问题，我们引入可视为由非精确辅助问题原理或非精确变度量前向-后向算法产生的下降方法。假设全局目标函数满足Kurdyka-Lojasiewicz不等式，我们证明所提算法的收敛性，弱化了先前工作中的假设。
引用
@article{arxiv.1508.02994,
title = {Auxiliary problem principle and inexact variable metric forward-backward algorithm for minimizing the sum of a differentiable function and a convex function},
author = {Jean-Philippe Chancelier},
journal= {arXiv preprint arXiv:1508.02994},
year = {2016}
}
备注
This paper has been withdrawn by the author due to crucial error in Lemma 4