Kimad: Adaptive Gradient Compression with Bandwidth Awareness

Author:

Xin Jihao1ORCID,Ilin Ivan1ORCID,Zhang Shunkang2ORCID,Canini Marco1ORCID,Richtárik Peter1ORCID

Affiliation:

1. KAUST, Thuwal, Saudi Arabia

2. HKUST, Hong Kong, China

Funder

King Abdullah University of Science and Technology Research Funding (KRF)

Publisher

ACM

Reference36 articles.

1. Ahmed M Abdelmoniem and Marco Canini. 2021. DC2: Delay-aware compression control for distributed machine learning. In INFOCOM. Ahmed M Abdelmoniem and Marco Canini. 2021. DC2: Delay-aware compression control for distributed machine learning. In INFOCOM.

2. Saurabh Agarwal , Hongyi Wang , Kangwook Lee , Shivaram Venkataraman , and Dimitris Papailiopoulos . 2021 . Accordion: Adaptive Gradient Communication via Critical Learning Regime Identification. In MLSys. Saurabh Agarwal, Hongyi Wang, Kangwook Lee, Shivaram Venkataraman, and Dimitris Papailiopoulos. 2021. Accordion: Adaptive Gradient Communication via Critical Learning Regime Identification. In MLSys.

3. Mohammadreza Alimohammadi Ilia Markov Elias Frantar and Dan Alistarh. 2022. L-GreCo: An Efficient and General Framework for Layerwise-Adaptive Gradient Compression. arXiv:2210.17357 [cs.LG] Mohammadreza Alimohammadi Ilia Markov Elias Frantar and Dan Alistarh. 2022. L-GreCo: An Efficient and General Framework for Layerwise-Adaptive Gradient Compression. arXiv:2210.17357 [cs.LG]

4. Dan Alistarh , Demjan Grubic , Jerry Li , Ryota Tomioka , and Milan Vojnovic . 2017 . QSGD: Communication-Efficient SGD via Gradient Quantization and Encoding. In NeurIPS. Dan Alistarh, Demjan Grubic, Jerry Li, Ryota Tomioka, and Milan Vojnovic. 2017. QSGD: Communication-Efficient SGD via Gradient Quantization and Encoding. In NeurIPS.

5. Dan Alistarh Torsten Hoefler Mikael Johansson Sarit Khirirat Nikola Konstantinov and Cédric Renggli. 2018. The Convergence of Sparsified Gradient Methods. In NeurIPS. Dan Alistarh Torsten Hoefler Mikael Johansson Sarit Khirirat Nikola Konstantinov and Cédric Renggli. 2018. The Convergence of Sparsified Gradient Methods. In NeurIPS.

Cited by 1 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. Accelerating Model Training in Multi-cluster Environments with Consumer-grade GPUs;Proceedings of the ACM SIGCOMM 2024 Conference;2024-08-04

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