Differentially Private Multi-Armed Bandits in the Shuffle Model

Jay Tenenbaum; Haim Kaplan; Yishay Mansour; Uri Stemmer

Differentially Private Multi-Armed Bandits in the Shuffle Model

Jay Tenenbaum, Haim Kaplan, Yishay Mansour, Uri Stemmer

School of Computer Science

Alphabet Inc.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

11 Scopus citations

Abstract

We give an (ε, δ)-differentially private algorithm for the multi-armed bandit (MAB) problem in the shuffle model with a distribution-dependent regret of^log_∆a^T +^k√^{log δ 1 log T} , and a distribution-independent regret_{O P a}∈[k]:_∆a>0 ε of O √kT log T +^k√^{log δ 1 log T} , where T is the number of rounds, ∆_a is the ε suboptimality gap of the arm a, and k is the total number of arms. Our upper bound almost matches the regret of the best known algorithms for the centralized model, and significantly outperforms the best known algorithm in the local model.

Original language	English
Title of host publication	Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021
Editors	Marc'Aurelio Ranzato, Alina Beygelzimer, Yann Dauphin, Percy S. Liang, Jenn Wortman Vaughan
Publisher	Neural information processing systems foundation
Pages	24956-24967
Number of pages	12
ISBN (Electronic)	9781713845393
State	Published - 2021
Event	35th Conference on Neural Information Processing Systems, NeurIPS 2021 - Virtual, Online Duration: 6 Dec 2021 → 14 Dec 2021

Publication series

Name	Advances in Neural Information Processing Systems
Volume	30
ISSN (Print)	1049-5258

Conference

Conference	35th Conference on Neural Information Processing Systems, NeurIPS 2021
City	Virtual, Online
Period	6/12/21 → 14/12/21

Funding

Funders	Funder number
Yandex Initiative for Machine Learning
European Research Council
Blavatnik Family Foundation
Israel Science Foundation	993/17,1595/19,1871/19
German-Israeli Foundation for Scientific Research and Development	1367/2017
Horizon 2020 Framework Programme	882396
Tel Aviv University

Cite this

Tenenbaum, J., Kaplan, H., Mansour, Y., & Stemmer, U. (2021). Differentially Private Multi-Armed Bandits in the Shuffle Model. In MA. Ranzato, A. Beygelzimer, Y. Dauphin, P. S. Liang, & J. Wortman Vaughan (Eds.), Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021 (pp. 24956-24967). (Advances in Neural Information Processing Systems; Vol. 30). Neural information processing systems foundation.

Tenenbaum, Jay ; Kaplan, Haim ; Mansour, Yishay et al. / Differentially Private Multi-Armed Bandits in the Shuffle Model. Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021. editor / Marc'Aurelio Ranzato ; Alina Beygelzimer ; Yann Dauphin ; Percy S. Liang ; Jenn Wortman Vaughan. Neural information processing systems foundation, 2021. pp. 24956-24967 (Advances in Neural Information Processing Systems).

@inproceedings{414d6f3da7a04904b85d091d0b81f880,

title = "Differentially Private Multi-Armed Bandits in the Shuffle Model",

abstract = "We give an (ε, δ)-differentially private algorithm for the multi-armed bandit (MAB) problem in the shuffle model with a distribution-dependent regret oflog∆aT +k√log δ 1 log T , and a distribution-independent regretO P a∈[k]:∆a>0 ε of O √kT log T +k√log δ 1 log T , where T is the number of rounds, ∆a is the ε suboptimality gap of the arm a, and k is the total number of arms. Our upper bound almost matches the regret of the best known algorithms for the centralized model, and significantly outperforms the best known algorithm in the local model.",

author = "Jay Tenenbaum and Haim Kaplan and Yishay Mansour and Uri Stemmer",

note = "Publisher Copyright: {\textcopyright} 2021 Neural information processing systems foundation. All rights reserved.; 35th Conference on Neural Information Processing Systems, NeurIPS 2021 ; Conference date: 06-12-2021 Through 14-12-2021",

year = "2021",

language = "אנגלית",

series = "Advances in Neural Information Processing Systems",

publisher = "Neural information processing systems foundation",

pages = "24956--24967",

editor = "Marc'Aurelio Ranzato and Alina Beygelzimer and Yann Dauphin and Liang, {Percy S.} and {Wortman Vaughan}, Jenn",

booktitle = "Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021",

}

Tenenbaum, J, Kaplan, H , Mansour, Y & Stemmer, U 2021, Differentially Private Multi-Armed Bandits in the Shuffle Model. in MA Ranzato, A Beygelzimer, Y Dauphin, PS Liang & J Wortman Vaughan (eds), Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021. Advances in Neural Information Processing Systems, vol. 30, Neural information processing systems foundation, pp. 24956-24967, 35th Conference on Neural Information Processing Systems, NeurIPS 2021, Virtual, Online, 6/12/21.

Differentially Private Multi-Armed Bandits in the Shuffle Model. / Tenenbaum, Jay; Kaplan, Haim ; Mansour, Yishay et al.
Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021. ed. / Marc'Aurelio Ranzato; Alina Beygelzimer; Yann Dauphin; Percy S. Liang; Jenn Wortman Vaughan. Neural information processing systems foundation, 2021. p. 24956-24967 (Advances in Neural Information Processing Systems; Vol. 30).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Differentially Private Multi-Armed Bandits in the Shuffle Model

AU - Tenenbaum, Jay

AU - Kaplan, Haim

AU - Mansour, Yishay

AU - Stemmer, Uri

PY - 2021

Y1 - 2021

N2 - We give an (ε, δ)-differentially private algorithm for the multi-armed bandit (MAB) problem in the shuffle model with a distribution-dependent regret oflog∆aT +k√log δ 1 log T , and a distribution-independent regretO P a∈[k]:∆a>0 ε of O √kT log T +k√log δ 1 log T , where T is the number of rounds, ∆a is the ε suboptimality gap of the arm a, and k is the total number of arms. Our upper bound almost matches the regret of the best known algorithms for the centralized model, and significantly outperforms the best known algorithm in the local model.

AB - We give an (ε, δ)-differentially private algorithm for the multi-armed bandit (MAB) problem in the shuffle model with a distribution-dependent regret oflog∆aT +k√log δ 1 log T , and a distribution-independent regretO P a∈[k]:∆a>0 ε of O √kT log T +k√log δ 1 log T , where T is the number of rounds, ∆a is the ε suboptimality gap of the arm a, and k is the total number of arms. Our upper bound almost matches the regret of the best known algorithms for the centralized model, and significantly outperforms the best known algorithm in the local model.

UR - http://www.scopus.com/inward/record.url?scp=85125032064&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85125032064

T3 - Advances in Neural Information Processing Systems

SP - 24956

EP - 24967

BT - Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021

A2 - Ranzato, Marc'Aurelio

A2 - Beygelzimer, Alina

A2 - Dauphin, Yann

A2 - Liang, Percy S.

A2 - Wortman Vaughan, Jenn

PB - Neural information processing systems foundation

T2 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021

Y2 - 6 December 2021 through 14 December 2021

ER -

Tenenbaum J, Kaplan H , Mansour Y , Stemmer U. Differentially Private Multi-Armed Bandits in the Shuffle Model. In Ranzato MA, Beygelzimer A, Dauphin Y, Liang PS, Wortman Vaughan J, editors, Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021. Neural information processing systems foundation. 2021. p. 24956-24967. (Advances in Neural Information Processing Systems).

Differentially Private Multi-Armed Bandits in the Shuffle Model

Abstract

Publication series

Conference

Funding

Other files and links

Fingerprint

Cite this