Principal Component Analysis in the local differential privacy model

被引:14
|
作者
Wang, Di [1 ]
Xu, Jinhui [1 ]
机构
[1] SUNY Buffalo, Dept Comp Sci & Engn, 338 Davis Hall, Buffalo, NY 14260 USA
关键词
Local differential privacy; Principal Component Analysis; Sparse PCA;
D O I
10.1016/j.tcs.2019.12.019
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
In this paper, we study the Principal Component Analysis (PCA) problem under the (distributed) non-interactive local differential privacy model. For the low dimensional case (i.e., p << n), we show the optimal rate of Theta(kp/n epsilon(2)) (omitting the eigenvalue terms) for the private minimax risk of the k-dimensional PCA using the squared subspace distance as the measurement, where n is the sample size and E is the privacy parameter. For the high dimensional (i.e., p >> n) row sparse case, we first give a lower bound of Omega(ks log p/n epsilon(2))) on the private minimax risk, where s is the underlying sparsity parameter. Then we provide an efficient algorithm to achieve the upper bound of O(s(2) log p/n epsilon(2)). Experiments on both synthetic and real world datasets confirm our theoretical guarantees. (C) 2019 Elsevier B.V. All rights reserved.
引用
收藏
页码:296 / 312
页数:17
相关论文
共 50 条
  • [41] Privacy at Scale: Local Differential Privacy in Practice
    Cormode, Graham
    Jha, Somesh
    Kulkarni, Tejas
    Li, Ninghui
    Srivastava, Divesh
    Wang, Tianhao
    SIGMOD'18: PROCEEDINGS OF THE 2018 INTERNATIONAL CONFERENCE ON MANAGEMENT OF DATA, 2018, : 1655 - 1658
  • [42] Behavior Sequence Mining Model Based on Local Differential Privacy
    Yan, Jianen
    Wang, Yan
    Li, Wenling
    IEEE ACCESS, 2020, 8 : 196086 - 196093
  • [43] BLENDER: Enabling Local Search with a Hybrid Differential Privacy Model
    Avent, Brendan
    Korolova, Aleksandra
    Zeber, David
    Hoyden, Torgeir
    Livshits, Benjamin
    PROCEEDINGS OF THE 26TH USENIX SECURITY SYMPOSIUM (USENIX SECURITY '17), 2017, : 747 - 764
  • [44] Privacy protection algorithm for the internet of vehicles based on local differential privacy and game model
    Han W.
    Cheng M.
    Lei M.
    Xu H.
    Yang Y.
    Qian L.
    Computers, Materials and Continua, 2020, 64 (02): : 1025 - 1038
  • [45] Privacy Protection Algorithm for the Internet of Vehicles Based on Local Differential Privacy and Game Model
    Han, Wenxi
    Cheng, Mingzhi
    Lei, Min
    Xii, Hanwen
    Yang, Yu
    Qian, Lei
    CMC-COMPUTERS MATERIALS & CONTINUA, 2020, 64 (02): : 1025 - 1038
  • [46] Fragility Assessment Model Based on Principal Component Analysis
    Wang, Zhenyan
    PROCEEDINGS OF THE 2018 8TH INTERNATIONAL CONFERENCE ON SOCIAL SCIENCE AND EDUCATION RESEARCH (SSER 2018), 2018, 238 : 50 - 55
  • [47] Structure space of model proteins: A principal component analysis
    Yahyanejad, M
    Kardar, M
    Tang, C
    JOURNAL OF CHEMICAL PHYSICS, 2003, 118 (09): : 4277 - 4284
  • [48] Dynamic kernel probabilistic principal component analysis model
    Institute of Automation, Jiangnan University, Wuxi 214122, China
    Qinghua Daxue Xuebao/Journal of Tsinghua University, 2008, 48 (SUPPL.): : 1824 - 1828
  • [49] High Dimensional Model Representation With Principal Component Analysis
    Hajikolaei, Kambiz Haji
    Wang, G. Gary
    JOURNAL OF MECHANICAL DESIGN, 2014, 136 (01)
  • [50] The improve model of the principal component analysis and its application
    Lu, ZH
    Li, JG
    PROCEEDINGS OF 2002 INTERNATIONAL CONFERENCE ON MANAGEMENT SCIENCE & ENGINEERING, VOLS I AND II, 2002, : 417 - 421