中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算机网络信息中心
A parameter communication optimization strategy for distributed machine learning in sensors

文献类型：期刊论文

作者Zhang, Jilin1,2,3,4,5; Tu, Hangdi1,2; Ren, Yongjian1,2; Wan, Jian1,2,4,5; Zhou, Li1,2; Li, Mingwei1,2; Wang, Jue6; Yu, Lifeng7,8; Zhao, Chang1,2; Zhang, Lei9
刊名Sensors
出版日期2017-10-01
卷号17期号:10页码:17
关键词Disturbed machine learning Sensors Dynamic synchronous parallel strategy (dsp) Parameter server (ps)
ISSN号1424-8220
DOI10.3390/s17102172
通讯作者Ren, yongjian(yongjian.ren@hdu.edu.cn)
英文摘要In order to utilize the distributed characteristic of sensors, distributed machine learning has become the mainstream approach, but the different computing capability of sensors and network delays greatly influence the accuracy and the convergence rate of the machine learning model. our paper describes a reasonable parameter communication optimization strategy to balance the training overhead and the communication overhead. we extend the fault tolerance of iterative-convergent machine learning algorithms and propose the dynamic finite fault tolerance (dfft). based on the dfft, we implement a parameter communication optimization strategy for distributed machine learning, named dynamic synchronous parallel strategy (dsp), which uses the performance monitoring model to dynamically adjust the parameter synchronization strategy between worker nodes and the parameter server (ps). this strategy makes full use of the computing power of each sensor, ensures the accuracy of the machine learning model, and avoids the situation that the model training is disturbed by any tasks unrelated to the sensors.
WOS关键词BODY AREA NETWORKS ; FRAMEWORK ; LIFETIME
WOS研究方向Chemistry ; Electrochemistry ; Instruments & Instrumentation
WOS类目Chemistry, Analytical ; Electrochemistry ; Instruments & Instrumentation
语种英语
WOS记录号WOS:000414931500014
出版者MDPI AG
URI标识http://www.irgrid.ac.cn/handle/1471x/2374255
专题计算机网络信息中心
通讯作者Ren, Yongjian
作者单位1.Hangzhou Dianzi Univ, Sch Comp Sci & Technol, Hangzhou 310018, Zhejiang, Peoples R China
2.Minist Educ, Key Lab Complex Syst Modeling & Simulat, Hangzhou 310018, Zhejiang, Peoples R China
3.Zhejiang Univ, Coll Elect Engn, Hangzhou 310058, Zhejiang, Peoples R China
4.Zhejiang Univ Sci & Technol, Sch Informat & Elect Engn, Hangzhou 310023, Zhejiang, Peoples R China
5.Zhejiang Prov Engn Ctr Media Data Cloud Proc & An, Hangzhou 310018, Zhejiang, Peoples R China
6.Chinese Acad Sci, Supercomp Ctr Comp Network Informat Ctr, Beijing 100190, Peoples R China
7.Hithink RoyalFlush Informat Network Co Ltd, Hangzhou 310023, Zhejiang, Peoples R China
8.Financial Informat Engn Technol Res Ctr Zhejiang, Hangzhou 310023, Zhejiang, Peoples R China
9.Beijing Univ Civil Engn & Architecture, Dept Comp Sci, Beijing 100044, Peoples R China
推荐引用方式
GB/T 7714		 Zhang, Jilin,Tu, Hangdi,Ren, Yongjian,et al. A parameter communication optimization strategy for distributed machine learning in sensors[J]. Sensors,2017,17(10):17.
APA Zhang, Jilin.,Tu, Hangdi.,Ren, Yongjian.,Wan, Jian.,Zhou, Li.,...&Zhang, Lei.(2017).A parameter communication optimization strategy for distributed machine learning in sensors.Sensors,17(10),17.
MLA Zhang, Jilin,et al."A parameter communication optimization strategy for distributed machine learning in sensors".Sensors 17.10(2017):17.

入库方式： iSwitch采集

来源：计算机网络信息中心

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。