Concentrated Differential Privacy for Bandits

被引:0
|
作者
Azize, Achraf [1 ]
Basu, Debabrota [1 ]
机构
[1] Univ Lille, Inria, CNRS, Cent Lille,CRIStAL,Equipe Scool, Lille, France
来源
IEEE CONFERENCE ON SAFE AND TRUSTWORTHY MACHINE LEARNING, SATML 2024 | 2024年
关键词
Differential Privacy; Multi-armed Bandits; Regret Analysis; Lower bounds;
D O I
10.1109/SaTML59370.2024.00013
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Bandits serve as the theoretical foundation of sequential learning and an algorithmic foundation of modern recommender systems. However, recommender systems often rely on user-sensitive data, making privacy a critical concern. This paper contributes to the understanding of Differential Privacy (DP) in bandits with a trusted centralised decision-maker, and especially the implications of ensuring zero Concentrated Differential Privacy (zCDP). First, we formalise and compare different adaptations of DP to bandits, depending on the considered input and the interaction protocol. Then, we propose three private algorithms, namely AdaC-UCB, AdaC-GOPE and AdaC-OFUL, for three bandit settings, namely finite-armed bandits, linear bandits, and linear contextual bandits. The three algorithms share a generic algorithmic blueprint, i.e. the Gaussian mechanism and adaptive episodes, to ensure a good privacy-utility trade-off. We analyse and upper bound the regret of these three algorithms. Our analysis shows that in all of these settings, the prices of imposing zCDP are (asymptotically) negligible in comparison with the regrets incurred oblivious to privacy. Next, we complement our regret upper bounds with the first minimax lower bounds on the regret of bandits with zCDP. To prove the lower bounds, we elaborate a new proof technique based on couplings and optimal transport. We conclude by experimentally validating our theoretical results for the three different settings of bandits.
引用
收藏
页码:78 / 109
页数:32
相关论文
共 50 条
  • [31] Image Pixelization with Differential Privacy
    Fan, Liyue
    DATA AND APPLICATIONS SECURITY AND PRIVACY XXXII, DBSEC 2018, 2018, 10980 : 148 - 162
  • [32] Honest Fraction Differential Privacy
    Taibi, Imane
    Ramon, Jan
    PROCEEDINGS OF THE 2024 ACM WORKSHOP ON INFORMATION HIDING AND MULTIMEDIA SECURITY, IH&MMSEC 2024, 2024, : 247 - 251
  • [33] Differential Privacy for Information Retrieval
    Yang, Grace Hui
    Zhang, Sicong
    ICTIR'17: PROCEEDINGS OF THE 2017 ACM SIGIR INTERNATIONAL CONFERENCE THEORY OF INFORMATION RETRIEVAL, 2017, : 325 - 326
  • [34] On the behavioral implications of differential privacy
    Gilboa-Freedman, Gail
    Smorodinsky, Rann
    THEORETICAL COMPUTER SCIENCE, 2020, 841 : 84 - 93
  • [35] The Composition Theorem for Differential Privacy
    Kairouz, Peter
    Oh, Sewoong
    Viswanath, Pramod
    IEEE TRANSACTIONS ON INFORMATION THEORY, 2017, 63 (06) : 4037 - 4049
  • [36] A survey on differential privacy and applications
    Xiong, Ping
    Zhu, Tian-Qing
    Wang, Xiao-Feng
    Jisuanji Xuebao/Chinese Journal of Computers, 2014, 37 (01): : 101 - 122
  • [37] Differential Privacy in the Local Setting
    Li, Ninghui
    IWSPA '18: PROCEEDINGS OF THE FOURTH ACM INTERNATIONAL WORKSHOP ON SECURITY AND PRIVACY ANALYTICS, 2018, : 42 - 42
  • [38] Prescriptive analytics with differential privacy
    Harikumar, Haripriya
    Rana, Santu
    Gupta, Sunil
    Nguyen, Thin
    Kaimal, Ramachandra
    Venkatesh, Svetha
    INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS, 2022, 13 (02) : 123 - 138
  • [39] A logical characterization of differential privacy
    Castiglioni, Valentina
    Chatzikokolakis, Konstantinos
    Palamidessi, Catuscia
    SCIENCE OF COMPUTER PROGRAMMING, 2020, 188
  • [40] Helmholtz machine with differential privacy
    Hu, Junying
    Sun, Kai
    Zhang, Hai
    INFORMATION SCIENCES, 2022, 613 : 888 - 903