.png){kind=icon}
返回
Communication-efficient and Byzantine-robust distributed learning with statistical guarantee
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2023.109312.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Communication efficiency and robustness are two major issues in modern distributed learning frame-works. This is due to the practical situations where some computing nodes may have limited commu-nication power or may behave adversarial behaviors. To address the two issues simultaneously, this pa-per develops two communication-efficient and robust distributed learning algorithms for convex prob-lems. Our motivation is based on surrogate likelihood framework and the median and trimmed mean operations. Particularly, the proposed algorithms are provably robust against Byzantine failures, and also achieve optimal statistical rates for strong convex losses and convex (non-smooth) penalties. For typical statistical models such as generalized linear models, our results show that statistical errors dominate op-timization errors in finite iterations. Simulated and real data experiments are conducted to demonstrate the numerical performance of our algorithms.(c) 2023 Elsevier Ltd. All rights reserved.
Keyword:
Distributed learning
Byzantine failure
Communication efficiency
Surrogate likelihood
Proximal algorithm
期刊
IF:
7.6
论文数:
1.3W
被引数:
4.5W
机构
引用论文
.png){kind=icon}
A computational avenue towards understanding and design of zwitterionic anti-biofouling materials
PREAI
分享收藏Energy-efficient load balancing scheme for two-tier communication in wireless sensor networks无线传感器网络中节能的两层通信负载均衡方案
PREAI
分享收藏
查看更多内容