基于虚拟罚项的非凸复合问题的递减步长法：从一般情形到凸正则约束情形
最优化与控制
2023-07-07 v1
摘要
在本文中，我们首先将[4]中提出的用于非凸约束问题的递减步长法推广到处理等式约束和复合类型的非光滑目标函数。然后我们考虑约束为凸且满足标准约束规范的特例，并表明在此设定下算法可大幅简化，从而降低每次迭代的计算负担。
引用
@article{arxiv.2007.16024,
title = {Diminishing Stepsize Methods for Nonconvex Composite Problems via Ghost Penalties: from the General to the Convex Regular Constrained Case},
author = {Francisco Facchinei and Vyacheskav Kungurtsevb and Lorenzo Lampariello and Gesualdo Scutari},
journal= {arXiv preprint arXiv:2007.16024},
year = {2023}
}
备注
arXiv admin note: text overlap with arXiv:1709.03384