Aperiodic Local SGD: Beyond Local SGD

Author:

Zhang Hao1,Wu Tingting1,Cheng Siyao1,Liu Jie2

Affiliation:

1. Harbin Institute of Technology, China

2. Harbin Institute of Technology (Shenzhen), China

Funder

National Natural Science Foundation of Heilongjiang Province

Programs for Science and Technology Development of Heilongjiang Province

Key Science Technology Specific Projects of Heilongjiang Province

National Key R&D Program of China

National Natural Science Foundation of China

Fundamental Research Funds for the Central Universities

Publisher

ACM

Reference41 articles.

1. Dan Alistarh , Demjan Grubic , Jerry Li , Ryota Tomioka , and Milan Vojnovic . 2017 . QSGD: Communication-efficient SGD via gradient quantization and encoding. In Advances in Neural Information Processing Systems. 1709–1720. Dan Alistarh, Demjan Grubic, Jerry Li, Ryota Tomioka, and Milan Vojnovic. 2017. QSGD: Communication-efficient SGD via gradient quantization and encoding. In Advances in Neural Information Processing Systems. 1709–1720.

2. Demystifying Parallel and Distributed Deep Learning

3. Luke N Darlow Elliot J Crowley Antreas Antoniou and Amos J Storkey. 2018. Cinic-10 is not imagenet or cifar-10. arXiv preprint arXiv:1810.03505(2018). Luke N Darlow Elliot J Crowley Antreas Antoniou and Amos J Storkey. 2018. Cinic-10 is not imagenet or cifar-10. arXiv preprint arXiv:1810.03505(2018).

4. Priya Goyal , Piotr Dollar , Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, and Kaiming He. 2017 . Accurate, Large Minibatch SGD : Training ImageNet in 1 Hour. In arXiv preprint arXiv:1706.02677. Priya Goyal, Piotr Dollar, Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, and Kaiming He. 2017. Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour. In arXiv preprint arXiv:1706.02677.

5. Local SGD with Periodic Averaging: Tighter Analysis and Adaptive Synchronization;Haddadpour Farzin;Advances in Neural Information Processing Systems,2019

Cited by 8 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. Federated Edge Learning with Blurred or Pseudo Data Sharing;Proceedings of the 53rd International Conference on Parallel Processing;2024-08-12

2. D2D-Assisted Adaptive Federated Learning in Energy-Constrained Edge Computing;Applied Sciences;2024-06-07

3. CC-FedAvg: Computationally Customized Federated Averaging;IEEE Internet of Things Journal;2024-02-01

4. Data-Augmentation-Based Federated Learning;IEEE Internet of Things Journal;2023-12-15

5. Decentralized Gradient Tracking with Fixed-Time Local Updates;2023 13th International Conference on Information Science and Technology (ICIST);2023-12-08

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