Fundamental Limits of Communication Efficiency for Model Aggregation in Distributed Learning: A Rate-Distortion Approach-Reference-Cited by-同舟云学术

登录注册会员服务联系我们

Fundamental Limits of Communication Efficiency for Model Aggregation in Distributed Learning: A Rate-Distortion Approach

Published:2023-01 Issue:1 Volume:71 Page:173-186
ISSN:0090-6778
Container-title:IEEE Transactions on Communications
language:
Short-container-title:IEEE Trans. Commun.

Author:

Zhang Naifu¹^ORCID,Tao Meixia¹^ORCID,Wang Jia¹^ORCID,Xu Fan¹^ORCID

Affiliation:

1. Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China

Funder

National Key R&D Project of China

NSF of China

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Electrical and Electronic Engineering

Link

http://xplorestaging.ieee.org/ielx7/26/10016736/09964262.pdf?arnumber=9964262

Reference28 articles.

1. SignSGD: Compressed optimisation for non-convex problems;bernstein;Proc Int Conf Mach Learn,2018

2. QSGD: Communication-efficient SGD via randomized quantization and encoding;alistarh;Proc Neural Inf Process Syst,2018

3. TernGrad: Ternary gradients to reduce communication in distributed deep learning;wen;Proc Neural Inf Process Syst,2017

4. Sparse Communication for Distributed Gradient Descent

5. Deep gradient compression: Reducing the communication bandwidth for distributed training;lin;Proc Int Conf Learn Represent,2018

Cited by 11 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Wireless federated learning for PR identification and analysis based on generalized information;Intelligent Systems with Applications;2024-09

2. Normalized flow networks and generalized information aided PR dynamic analysis;Intelligent Systems with Applications;2024-09

3. Long-short term memory networks aided fault detection of power facilities;Intelligent Systems with Applications;2024-06

4. Knowledge graph learning algorithm based on deep convolutional networks;Intelligent Systems with Applications;2024-06

5. Secrecy-Driven Energy Minimization in Federated-Learning-Assisted Marine Digital Twin Networks;IEEE Internet of Things Journal;2024-02-01

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号京ICP备18003416号-3