Seeking Consensus on Subspaces in Federated Principal Component Analysis

被引:0
|
作者
Wang, Lei [1 ]
Liu, Xin [2 ,3 ]
Zhang, Yin [4 ]
机构
[1] Hong Kong Polytech Univ, Dept Appl Math, Hong Kong, Peoples R China
[2] Chinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
[3] Univ Chinese Acad Sci, Beijing, Peoples R China
[4] Chinese Univ Hong Kong, Shenzhen, Peoples R China
基金
中国国家自然科学基金;
关键词
Alternating direction method of multipliers; Federated learning; Principal component analysis; Orthogonality constraints; SIMULTANEOUS-ITERATION; OPTIMIZATION PROBLEMS; FRAMEWORK; ALGORITHM; SVD;
D O I
10.1007/s10957-024-02523-1
中图分类号
C93 [管理学]; O22 [运筹学];
学科分类号
070105 ; 12 ; 1201 ; 1202 ; 120202 ;
摘要
In this paper, we develop an algorithm for federated principal component analysis (PCA) with emphases on both communication efficiency and data privacy. Generally speaking, federated PCA algorithms based on direct adaptations of classic iterative methods, such as simultaneous subspace iterations, are unable to preserve data privacy, while algorithms based on variable-splitting and consensus-seeking, such as alternating direction methods of multipliers (ADMM), lack in communication-efficiency. In this work, we propose a novel consensus-seeking formulation by equalizing subspaces spanned by splitting variables instead of equalizing variables themselves, thus greatly relaxing feasibility restrictions and allowing much faster convergence. Then we develop an ADMM-like algorithm with several special features to make it practically efficient, including a low-rank multiplier formula and techniques for treating subproblems. We establish that the proposed algorithm can better protect data privacy than classic methods adapted to the federated PCA setting. We derive convergence results, including a worst-case complexity estimate, for the proposed ADMM-like algorithm in the presence of the nonlinear equality constraints. Extensive empirical results are presented to show that the new algorithm, while enhancing data privacy, requires far fewer rounds of communication than existing peer algorithms for federated PCA.
引用
收藏
页码:529 / 561
页数:33
相关论文
共 50 条
  • [1] Federated Supervised Principal Component Analysis
    Briguglio, William
    Yousef, Waleed A.
    Traore, Issa
    Mamun, Mohammad
    IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2024, 19 : 646 - 660
  • [2] Vertical Federated Principal Component Analysis on Feature-Wise Distributed Data
    Cheung, Yiu-Ming
    Lou, Jian
    Yu, Feng
    WEB INFORMATION SYSTEMS ENGINEERING - WISE 2021, PT I, 2021, 13080 : 173 - 188
  • [4] AN ALGORITHM FOR THE PRINCIPAL COMPONENT ANALYSIS OF LARGE DATA SETS
    Halko, Nathan
    Martinsson, Per-Gunnar
    Shkolnisky, Yoel
    Tygert, Mark
    SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2011, 33 (05) : 2580 - 2594
  • [5] Adaptive consensus principal component analysis for on-line batch process monitoring
    Lee, DS
    Vanrolleghem, PA
    ENVIRONMENTAL MONITORING AND ASSESSMENT, 2004, 92 (1-3) : 119 - 135
  • [6] Adaptive Consensus Principal Component Analysis for On-Line Batch Process Monitoring
    Dae Sung Lee
    Peter A. Vanrolleghem
    Environmental Monitoring and Assessment, 2004, 92 : 119 - 135
  • [7] Bilinear Probabilistic Principal Component Analysis
    Zhao, Jianhua
    Yu, Philip L. H.
    Kwok, James T.
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2012, 23 (03) : 492 - 503
  • [8] Accelerating Wireless Federated Learning via Nesterov's Momentum and Distributed Principal Component Analysis
    Dong, Yanjie
    Wang, Luya
    Wang, Jia
    Hu, Xiping
    Zhang, Haijun
    Yu, Fei Richard
    Leung, Victor C. M.
    IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2024, 23 (06) : 5938 - 5952
  • [9] Characterising particle packings by principal component analysis
    Feng, Y. T.
    Zhao, Tingting
    Wang, Min
    Owen, D. R. J.
    COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2018, 340 : 70 - 89
  • [10] Robust kernel principal component analysis and classification
    Debruyne, Michiel
    Verdonck, Tim
    ADVANCES IN DATA ANALYSIS AND CLASSIFICATION, 2010, 4 (2-3) : 151 - 167