Better Algorithms for Stochastic Bandits with Adversarial Corruptions

Anupam Gupta; Tomer Koren; Kunal Talwar

Better Algorithms for Stochastic Bandits with Adversarial Corruptions

Anupam Gupta, Tomer Koren, Kunal Talwar

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

Abstract

We study the stochastic multi-armed bandits problem in the presence of adversarial corruption. We present a new algorithm for this problem whose regret is nearly optimal, substantially improving upon previous work. Our algorithm is agnostic to the level of adversarial contamination and can tolerate a significant amount of corruption with virtually no degradation in performance.

Original language	English
Title of host publication	Proceedings of the Thirty-Second Conference on Learning Theory
Editors	Alina Beygelzimer, Daniel Hsu
Pages	1562-1578
Number of pages	17
State	Published - 2019
Externally published	Yes
Event	32nd Annual Conference on Learning Theory, COLT 2019 - Phoenix, United States Duration: 25 Jun 2019 → 28 Jun 2019 Conference number: 32

Publication series

Name	Proceedings of Machine Learning Research
Publisher	PMLR
Volume	99

Conference

Conference	32nd Annual Conference on Learning Theory, COLT 2019
Abbreviated title	COLT 2019
Country/Territory	United States
City	Phoenix
Period	25/06/19 → 28/06/19

Access to Document

Cite this

@inproceedings{7cca437dd2d54e718950e77f0328c120,

title = "Better Algorithms for Stochastic Bandits with Adversarial Corruptions",

abstract = "We study the stochastic multi-armed bandits problem in the presence of adversarial corruption. We present a new algorithm for this problem whose regret is nearly optimal, substantially improving upon previous work. Our algorithm is agnostic to the level of adversarial contamination and can tolerate a significant amount of corruption with virtually no degradation in performance.",

author = "Anupam Gupta and Tomer Koren and Kunal Talwar",

note = "DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.; 32nd Annual Conference on Learning Theory, COLT 2019 ; Conference date: 25-06-2019 Through 28-06-2019",

year = "2019",

language = "الإنجليزيّة",

series = "Proceedings of Machine Learning Research",

publisher = "PMLR",

pages = "1562--1578",

editor = "Alina Beygelzimer and Daniel Hsu",

booktitle = "Proceedings of the Thirty-Second Conference on Learning Theory",

}

Gupta, A, Koren, T & Talwar, K 2019, Better Algorithms for Stochastic Bandits with Adversarial Corruptions. in A Beygelzimer & D Hsu (eds), Proceedings of the Thirty-Second Conference on Learning Theory. Proceedings of Machine Learning Research, vol. 99, pp. 1562-1578, 32nd Annual Conference on Learning Theory, COLT 2019, Phoenix, Arizona, United States, 25/06/19. <https://dblp.org/db/conf/colt/colt2019.html#GuptaKT19>

Better Algorithms for Stochastic Bandits with Adversarial Corruptions. / Gupta, Anupam; Koren, Tomer; Talwar, Kunal.
Proceedings of the Thirty-Second Conference on Learning Theory. ed. / Alina Beygelzimer; Daniel Hsu. 2019. p. 1562-1578 (Proceedings of Machine Learning Research; Vol. 99).

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

TY - GEN

T1 - Better Algorithms for Stochastic Bandits with Adversarial Corruptions

AU - Gupta, Anupam

AU - Koren, Tomer

AU - Talwar, Kunal

N1 - Conference code: 32

PY - 2019

Y1 - 2019

N2 - We study the stochastic multi-armed bandits problem in the presence of adversarial corruption. We present a new algorithm for this problem whose regret is nearly optimal, substantially improving upon previous work. Our algorithm is agnostic to the level of adversarial contamination and can tolerate a significant amount of corruption with virtually no degradation in performance.

AB - We study the stochastic multi-armed bandits problem in the presence of adversarial corruption. We present a new algorithm for this problem whose regret is nearly optimal, substantially improving upon previous work. Our algorithm is agnostic to the level of adversarial contamination and can tolerate a significant amount of corruption with virtually no degradation in performance.

M3 - منشور من مؤتمر

T3 - Proceedings of Machine Learning Research

SP - 1562

EP - 1578

BT - Proceedings of the Thirty-Second Conference on Learning Theory

A2 - Beygelzimer, Alina

A2 - Hsu, Daniel

T2 - 32nd Annual Conference on Learning Theory, COLT 2019

Y2 - 25 June 2019 through 28 June 2019

ER -

Better Algorithms for Stochastic Bandits with Adversarial Corruptions

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this