Top-k Discovery Under Local Differential Privacy: An Adaptive Sampling Approach

被引:0
作者
Du, Rong [1 ]
Ye, Qingqing [1 ]
Fu, Yue [1 ]
Hu, Haibo [1 ]
Huang, Kai [2 ]
机构
[1] Hong Kong Polytech Univ, Dept Elect & Elect Engn, Hong Kong, Peoples R China
[2] Macau Univ Sci & Technol, Sch Comp Sci & Engn, Macau, Peoples R China
基金
中国国家自然科学基金;
关键词
Frequency estimation; Estimation; Differential privacy; Privacy; Radio spectrum management; Data collection; Time-frequency analysis; Random variables; Protocols; Probability distribution; Local differential privacy; multi-armed bandit; top- k Estimation; DATA PUBLICATION; BOUNDS;
D O I
10.1109/TDSC.2024.3471923
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Local differential privacy (LDP) is a promising privacy model for data collection that protects sensitive information of individuals. However, applying LDP to top-k estimation in set-valued data (e.g., identifying most frequent k items) may yield poor results for small and sparse datasets due to high sensitivity and heavy perturbation. To address this, we propose an adaptive approach that frames the problem as a multi-armed bandit (MAB) problem, in which the decision-maker selects actions based on information collected from previous rounds to maximize the total reward over time. Inspired by this, we present two adaptive sampling schemes based on MAB: ARBS for identifying top-k items and ARBSF for both top-k item discovery and frequency estimation on these items. Furthermore, to address the potential long delay of multi-round collection, we propose an optimization technique to reduce the time complexity. Both theoretical and empirical results show that our adaptive sampling schemes significantly outperform existing alternatives.
引用
收藏
页码:1763 / 1780
页数:18
相关论文
共 50 条
[31]   Building Quadtrees for Spatial Data Under Local Differential Privacy [J].
Alptekin, Ece ;
Gursoy, M. Emre .
DATA AND APPLICATIONS SECURITY AND PRIVACY XXXVII, DBSEC 2023, 2023, 13942 :22-39
[32]   Collaborative Sampling for Partial Multi-Dimensional Value Collection Under Local Differential Privacy [J].
Qian, Qiuyu ;
Ye, Qingqing ;
Hu, Haibo ;
Huang, Kai ;
Chan, Tom Tak-Lam ;
Li, Jin .
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2023, 18 :3948-3961
[33]   Local Differential Privacy in Graph Neural Networks: a Reconstruction Approach [J].
Bhaila, Karuna ;
Huang, Wen ;
Wu, Yongkai ;
Wu, Xintao .
PROCEEDINGS OF THE 2024 SIAM INTERNATIONAL CONFERENCE ON DATA MINING, SDM, 2024, :1-9
[34]   Secure Hot Path Crowdsourcing With Local Differential Privacy Under Fog Computing Architecture [J].
Yang, Mengmeng ;
Tjuawinata, Ivan ;
Lam, Kwok Yan ;
Zhao, Jun ;
Sun, Lin .
IEEE TRANSACTIONS ON SERVICES COMPUTING, 2022, 15 (04) :2188-2201
[35]   Fisher information under local differential privacy [J].
Barnes L.P. ;
Chen W.-N. ;
Özgür A. .
IEEE Journal on Selected Areas in Information Theory, 2020, 1 (03) :645-659
[36]   On the Security of Distributed Multi-Agent K-Means Clustering With Local Differential Privacy [J].
Shi, Congcong ;
Huang, Xiuli ;
Yu, Pengfei .
IEEE ACCESS, 2024, 12 :124751-124763
[37]   Optimizing Partial Area Under the Top-k Curve: Theory and Practice [J].
Wang, Zitai ;
Xu, Qianqian ;
Yang, Zhiyong ;
He, Yuan ;
Cao, Xiaochun ;
Huang, Qingming .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (04) :5053-5069
[38]   Statistical Quantification of Differential Privacy: A Local Approach [J].
Askin, Oender ;
Kutta, Tim ;
Dette, Holger .
43RD IEEE SYMPOSIUM ON SECURITY AND PRIVACY (SP 2022), 2022, :402-421
[39]   Random Sampling Plus Fake Data: Multidimensional Frequency Estimates With Local Differential Privacy [J].
Arcolezi, Heber H. ;
Couchot, Jean-Francois ;
Al Bouna, Bechara ;
Xiao, Xiaokui .
PROCEEDINGS OF THE 30TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT, CIKM 2021, 2021, :47-57
[40]   Privacy-Preserving Approximate Top-k Nearest Keyword Queries over Encrypted Graphs [J].
Shen, Meng ;
Wang, Minghui ;
Xu, Ke ;
Zhu, Liehuang .
2021 IEEE/ACM 29TH INTERNATIONAL SYMPOSIUM ON QUALITY OF SERVICE (IWQOS), 2021,