Proximal Methods Avoid Active Strict Saddles of Weakly Convex Functions
Proximal Methods Avoid Active Strict Saddles of Weakly Convex Functions
复制标题
近端方法避免弱凸函数的主动严格鞍点
DOI:
10.1007/s10208-021-09516-w
复制
发表时间:
2021
影响因子:
3
通讯作者:
Drusvyatskiy, Dmitriy
中科院分区:
文献类型:
--
作者:
Davis, Damek;Drusvyatskiy, Dmitriy
We introduce a geometrically transparent strict saddle property for nonsmooth functions. This property guarantees that simple proximal algorithms on weakly convex problems converge only to local minimizers, when randomly initialized. We argue that the strict saddle property may be a realistic assumption in applications, since it provably holds for generic semi-algebraic optimization problems.
登录
查看更多内容
DOI:
--
发表时间:
2019
期刊:
影响因子:
--
作者:
Dmitrii Avdiukhin;Chi Jin
通讯作者:
Chi Jin
影响因子:
8.7
作者:
P. Albano;P. Cannarsa
通讯作者:
P. Cannarsa
影响因子:
3.1
作者:
Duchi, John C.;Ruan, Feng
通讯作者:
Ruan, Feng
影响因子:
2.1
作者:
Beck, Amir;Teboulle, Marc
通讯作者:
Teboulle, Marc
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