Optimization and Control

[31] RSA: Byzantine-robust stochastic aggregation methods for distributed learning from heterogeneous datasets

Liping Li Wei Xu Tianyi Chen Georgios Giannakis Qing Ling

Optimization and Control mathscidoc:2004.27012

AAAI, 2019
[ Download ] [ 2020-04-27 01:45:30 uploaded by qinglingustc ] [ 957 downloads ] [ 0 comments ] [ Abstract ] [ Full ]
Please log in for comment!
 

[32] Primal-dual optimization algorithms over Riemannian manifolds: an iteration complexity analysis

Junyu Zhang University of Minnesota Shiqian Ma University of California, Davis Shuzhong Zhang University of Minnesota

Optimization and Control mathscidoc:2004.27011

Mathematical Programming, 2019.8
[ Download ] [ 2020-04-27 01:44:35 uploaded by concor123 ] [ 855 downloads ] [ 0 comments ] [ Abstract ] [ Full ]
Please log in for comment!
 

[33] Can primal methods outperform primal-dual methods in decentralized dynamic optimization?

Kun Yuan UCLA Wei Xu USTC Qing Ling SYSU

Optimization and Control mathscidoc:2004.27010

IEEE Transactions on Signal Processing, 2020
[ Download ] [ 2020-04-27 01:42:40 uploaded by qinglingustc ] [ 590 downloads ] [ 0 comments ] [ Abstract ] [ Full ]
Please log in for comment!
 

[34] Federated variance-reduced stochastic gradient descent with robustness to Byzantine attacks

Zhaoxian Wu SYSU Qing Ling SYSU Tianyi Chen RPI Georgios Giannakis UMN

Optimization and Control mathscidoc:2004.27009

IEEE Transactions on Signal Processing, 2020
[ Download ] [ 2020-04-27 01:41:19 uploaded by qinglingustc ] [ 688 downloads ] [ 0 comments ] [ Abstract ] [ Full ]
Please log in for comment!
 

[35] Communication-censored distributed stochastic gradient descent

Weiyu Li USTC Tianyi Chen RPI Liping Li USTC Zhaoxian Wu SYSU Qing Ling SYSU

Optimization and Control mathscidoc:2004.27008

IEEE Transactions on Neural Networks and Learning Systems, 2020
[ Download ] [ 2020-04-27 01:39:38 uploaded by qinglingustc ] [ 559 downloads ] [ 0 comments ] [ Abstract ] [ Full ]
Please log in for comment!
 

Show all 3 5 10 25 papers per page.
Sort by time views
 
Contact us: office-iccm@tsinghua.edu.cn | Copyright Reserved