MinMax Sampling: A Near-optimal Global Summary for Aggregation in the Wide Area

被引:5
|
作者
Zhao, Yikai [1 ,2 ,3 ]
Zhang, Yinda [1 ,2 ,3 ]
Li, Yuanpeng [1 ,2 ,3 ]
Zhou, Yi [1 ,2 ,3 ]
Chen, Chunhui [1 ,5 ]
Yang, Tong [1 ,2 ,3 ,4 ]
Cui, Bin [1 ,2 ,3 ]
机构
[1] Peking Univ, Beijing, Peoples R China
[2] Peking Univ, Sch Comp Sci, Beijing, Peoples R China
[3] Peking Univ, Natl Engn Lab Big Data Anal Technol & Applicat, Beijing, Peoples R China
[4] Peng Cheng Lab, Shenzhen, Peoples R China
[5] Peking Univ, Sch Software & Microelect, Beijing, Peoples R China
来源
PROCEEDINGS OF THE 2022 INTERNATIONAL CONFERENCE ON MANAGEMENT OF DATA (SIGMOD '22) | 2022年
基金
中国国家自然科学基金;
关键词
Sampling; Wide-Area Network; Federated Learning; FREQUENT;
D O I
10.1145/3514221.3526160
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Nowadays, wide-area data analyses are pervasive with emerging geo-distributed systems. These analyses often need to do the global aggregation in the wide area. Since scarce and variable WAN bandwidth may degrade the aggregation performance, it is highly desired to design a communication scheme for global aggregation in WAN. Unfortunately, no existing algorithm can meet the three design requirements of communication schemes: fast computation, adaptive transmission, and accurate aggregation. In this paper, we propose MinMax Sampling, a fast, adaptive, and accurate communication scheme for global aggregation in WAN. We first focus on the accuracy and design a scheme, namely MinMax(opt), to achieve optimal accuracy. However, MinMax(opt) does not meet the other two requirements: fast computation and adaptive transmission. Based on MinMax(opt), we propose MinMax(adp), which trades little accuracy for the other two requirements. We evaluate MinMax(adp) with three applications: federated learning, distributed state aggregation, and hierarchical aggregation. Our experimental results show that MinMax(adp) is superior to existing algorithms (8.44x better accuracy on average) in all three applications. The source codes of MinMax Sampling are available at Github [1].
引用
收藏
页码:744 / 758
页数:15
相关论文
共 50 条
  • [31] Optimal and near-optimal global register allocation using 0-1 integer programming
    Univ of California, Davis, United States
    Software Pract Exper, 8 (929-965):
  • [32] Optimal and near-optimal global register allocation using 0-1 integer programming
    Goodwin, DW
    Wilken, KD
    SOFTWARE-PRACTICE & EXPERIENCE, 1996, 26 (08): : 929 - 965
  • [33] Leaf area index simulation in soybean grown under near-optimal conditions
    Setiyono, T. D.
    Weiss, A.
    Specht, J. E.
    Cassman, K. G.
    Dobermann, A.
    FIELD CROPS RESEARCH, 2008, 108 (01) : 82 - 92
  • [34] Near-optimal solutions of convex semi-infinite programs via targeted sampling
    Souvik Das
    Ashwin Aravind
    Ashish Cherukuri
    Debasish Chatterjee
    Annals of Operations Research, 2022, 318 : 129 - 146
  • [35] Near-Optimal Thompson Sampling-based Algorithms for Differentially Private Stochastic Bandits
    Hu, Bingshan
    Hegde, Nidhi
    UNCERTAINTY IN ARTIFICIAL INTELLIGENCE, VOL 180, 2022, 180 : 844 - +
  • [36] Design of Optimal and Near-Optimal Enterprise-Wide Supply Networks for Multiple Products in the Process Industry
    Fan, L. T.
    Kim, Young
    Yun, Choamun
    Park, Seung Bin
    Park, Sunwon
    Bertok, Botond
    Friedler, Ferenc
    INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 2009, 48 (04) : 2003 - 2008
  • [37] Layered Gibbs Sampling Algorithm For Near-Optimal Detection in Large-MIMO Systems
    Mandloi, Manish
    Bhatia, Vimal
    2017 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE (WCNC), 2017,
  • [38] A Layered Iterative Sampling Algorithm for Near-Optimal Detection in Uplink Massive MIMO Systems
    Zhang, Nutian
    Fei, Chao
    Zhang, Zhenbing
    Hu, Jianhao
    Chen, Jienan
    2017 9TH INTERNATIONAL CONFERENCE ON WIRELESS COMMUNICATIONS AND SIGNAL PROCESSING (WCSP), 2017,
  • [39] Near-optimal solutions of convex semi-infinite programs via targeted sampling
    Das, Souvik
    Aravind, Ashwin
    Cherukuri, Ashish
    Chatterjee, Debasish
    ANNALS OF OPERATIONS RESEARCH, 2022, 318 (01) : 129 - 146
  • [40] Provably near-optimal sampling-based policies for stochastic inventory control models
    Levi, Retsef
    Roundy, Robin O.
    Shmoys, David B.
    MATHEMATICS OF OPERATIONS RESEARCH, 2007, 32 (04) : 821 - 839