This website requires JavaScript.

Federated Learning via Inexact ADMM

Shenglong ZhouGeoffrey Ye Li
Apr 2022
摘要
One of the crucial issues in federated learning is how to develop efficientoptimization algorithms. Most of the current ones require full devicesparticipation and/or impose strong assumptions for convergence. Different fromthe widely-used gradient descent-based algorithms, this paper develops aninexact alternating direction method of multipliers (ADMM), which is bothcomputation and communication-efficient, capable of combating the stragglers'effect, and convergent under mild conditions.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答