A Differentially Private Approach for Budgeted Combinatorial Multi-Armed Bandits

被引:0
|
作者
Wang, Hengzhi [1 ]
Cui, Laizhong [1 ]
Wang, En [2 ,3 ]
Liu, Jiangchuan [4 ]
机构
[1] Shenzhen Univ, Coll Comp Sci & Software Engn, Shenzhen 518060, Peoples R China
[2] Jilin Univ, Dept Comp Sci & Technol, Changchun 130012, Peoples R China
[3] Jilin Univ, Key Lab Symbol Computat & Knowledge Engn, Minist Educ, Changchun 130012, Peoples R China
[4] Simon Fraser Univ, Sch Comp Sci, Burnaby, BC V5A 1S6, Canada
基金
中国国家自然科学基金;
关键词
Combinatorial multi-armed bandits; differential privacy; online learning; truthfulness;
D O I
10.1109/TDSC.2024.3401836
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
As a fundamental tool for sequential decision-making, the Combinatorial Multi-Armed Bandits model (CMAB) has been extensively analyzed and applied in various online applications. However, the privacy concerns in budgeted CMAB are rarely investigated thus far. Few bandit algorithms have adequately addressed the privacy-preserving budgeted CMAB setting. Motivated by this, we study this setting using differential privacy as the formal measure of privacy. In this setting, playing an arm yields both a random reward and a random cost, and these values are kept private. In addition, multiple arms can be played in each round. The objective of the decision-maker is to minimize regret while subject to a budget constraint on the cumulative cost of all played arms. We demonstrate an exploration-exploitation-balanced bandit policy, which preserves the privacy of both rewards and costs under budgeted CMAB settings. This policy is proven differentially private and achieves an upper bound on regret. Furthermore, to provide incentives for the differentially private bandit policy so as to ensure that the reported costs are truthful, we introduce the concept of truthfulness and incorporate a payment mechanism that has been proven to be $\sigma$sigma-truthful. Numerical simulations based on multiple real-world datasets validate the theoretical findings and demonstrate the effectiveness of our policy compared to state-of-the-art policies.
引用
收藏
页码:424 / 439
页数:16
相关论文
共 50 条
  • [41] On Optimal Foraging and Multi-armed Bandits
    Srivastava, Vaibhav
    Reverdy, Paul
    Leonard, Naomi E.
    2013 51ST ANNUAL ALLERTON CONFERENCE ON COMMUNICATION, CONTROL, AND COMPUTING (ALLERTON), 2013, : 494 - 499
  • [42] Active Learning in Multi-armed Bandits
    Antos, Andras
    Grover, Varun
    Szepesvari, Csaba
    ALGORITHMIC LEARNING THEORY, PROCEEDINGS, 2008, 5254 : 287 - +
  • [43] Multi-Armed Bandits with Cost Subsidy
    Sinha, Deeksha
    Sankararama, Karthik Abinav
    Kazerouni, Abbas
    Avadhanula, Vashist
    24TH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS (AISTATS), 2021, 130
  • [44] Multi-Armed Bandits With Correlated Arms
    Gupta, Samarth
    Chaudhari, Shreyas
    Joshi, Gauri
    Yagan, Osman
    IEEE TRANSACTIONS ON INFORMATION THEORY, 2021, 67 (10) : 6711 - 6732
  • [45] Batched Multi-armed Bandits Problem
    Gao, Zijun
    Han, Yanjun
    Ren, Zhimei
    Zhou, Zhengqing
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019, 32
  • [46] Are Multi-Armed Bandits Susceptible to Peeking?
    Loecher, Markus
    ZAGREB INTERNATIONAL REVIEW OF ECONOMICS & BUSINESS, 2018, 21 (01): : 95 - 104
  • [47] Secure Outsourcing of Multi-Armed Bandits
    Ciucanu, Radu
    Lafourcade, Pascal
    Lombard-Platet, Marius
    Soare, Marta
    2020 IEEE 19TH INTERNATIONAL CONFERENCE ON TRUST, SECURITY AND PRIVACY IN COMPUTING AND COMMUNICATIONS (TRUSTCOM 2020), 2020, : 202 - 209
  • [48] Decentralized Exploration in Multi-Armed Bandits
    Feraud, Raphael
    Alami, Reda
    Laroche, Romain
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 97, 2019, 97
  • [49] Multi-armed bandits with episode context
    Rosin, Christopher D.
    ANNALS OF MATHEMATICS AND ARTIFICIAL INTELLIGENCE, 2011, 61 (03) : 203 - 230
  • [50] Introduction to Multi-Armed Bandits Preface
    Slivkins, Aleksandrs
    FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2019, 12 (1-2): : 1 - 286