Concurrent bandits and cognitive radio networks

Orly Avner; Shie Mannor

doi:https://doi.org/10.1007/978-3-662-44848-9_5

Concurrent bandits and cognitive radio networks

Orly Avner, Shie Mannor

Technion - Israel Institute of Technology

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

Abstract

We consider the problem of multiple users targeting the arms of a single multi-armed stochastic bandit. The motivation for this problem comes from cognitive radio networks, where selfish users need to coexist without any side communication between them, implicit cooperation or common control. Even the number of users may be unknown and can vary as users join or leave the network. We propose an algorithm that combines an ε-greedy learning rule with a collision avoidance mechanism. We analyze its regret with respect to the system-wide optimum and show that sub-linear regret can be obtained in this setting. Experiments show dramatic improvement compared to other algorithms for this setting.

Original language	English
Title of host publication	Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2014, Proceedings
Pages	66-81
Number of pages	16
Edition	PART 1
DOIs	https://doi.org/10.1007/978-3-662-44848-9_5
State	Published - 2014
Event	European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2014 - Nancy, France Duration: 15 Sep 2014 → 19 Sep 2014

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 1
Volume	8724 LNAI

Conference

Conference	European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2014
Country/Territory	France
City	Nancy
Period	15/09/14 → 19/09/14

Keywords

Bandits
Epsilon-greedy
Multi-user

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

https://doi.org/10.1007/978-3-662-44848-9_5

Cite this

Avner, O., & Mannor, S. (2014). Concurrent bandits and cognitive radio networks. In Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2014, Proceedings (PART 1 ed., pp. 66-81). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8724 LNAI, No. PART 1). https://doi.org/10.1007/978-3-662-44848-9_5

@inproceedings{21b4e2441861479190b1f1c87029ce7a,

title = "Concurrent bandits and cognitive radio networks",

abstract = "We consider the problem of multiple users targeting the arms of a single multi-armed stochastic bandit. The motivation for this problem comes from cognitive radio networks, where selfish users need to coexist without any side communication between them, implicit cooperation or common control. Even the number of users may be unknown and can vary as users join or leave the network. We propose an algorithm that combines an ε-greedy learning rule with a collision avoidance mechanism. We analyze its regret with respect to the system-wide optimum and show that sub-linear regret can be obtained in this setting. Experiments show dramatic improvement compared to other algorithms for this setting.",

keywords = "Bandits, Epsilon-greedy, Multi-user",

author = "Orly Avner and Shie Mannor",

year = "2014",

doi = "https://doi.org/10.1007/978-3-662-44848-9_5",

language = "الإنجليزيّة",

isbn = "9783662448472",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 1",

pages = "66--81",

booktitle = "Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2014, Proceedings",

edition = "PART 1",

note = "European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2014 ; Conference date: 15-09-2014 Through 19-09-2014",

}

Avner, O & Mannor, S 2014, Concurrent bandits and cognitive radio networks. in Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2014, Proceedings. PART 1 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 8724 LNAI, pp. 66-81, European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2014, Nancy, France, 15/09/14. https://doi.org/10.1007/978-3-662-44848-9_5

Concurrent bandits and cognitive radio networks. / Avner, Orly; Mannor, Shie.
Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2014, Proceedings. PART 1. ed. 2014. p. 66-81 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8724 LNAI, No. PART 1).

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

TY - GEN

T1 - Concurrent bandits and cognitive radio networks

AU - Avner, Orly

AU - Mannor, Shie

PY - 2014

Y1 - 2014

N2 - We consider the problem of multiple users targeting the arms of a single multi-armed stochastic bandit. The motivation for this problem comes from cognitive radio networks, where selfish users need to coexist without any side communication between them, implicit cooperation or common control. Even the number of users may be unknown and can vary as users join or leave the network. We propose an algorithm that combines an ε-greedy learning rule with a collision avoidance mechanism. We analyze its regret with respect to the system-wide optimum and show that sub-linear regret can be obtained in this setting. Experiments show dramatic improvement compared to other algorithms for this setting.

AB - We consider the problem of multiple users targeting the arms of a single multi-armed stochastic bandit. The motivation for this problem comes from cognitive radio networks, where selfish users need to coexist without any side communication between them, implicit cooperation or common control. Even the number of users may be unknown and can vary as users join or leave the network. We propose an algorithm that combines an ε-greedy learning rule with a collision avoidance mechanism. We analyze its regret with respect to the system-wide optimum and show that sub-linear regret can be obtained in this setting. Experiments show dramatic improvement compared to other algorithms for this setting.

KW - Bandits

KW - Epsilon-greedy

KW - Multi-user

UR - http://www.scopus.com/inward/record.url?scp=84907017993&partnerID=8YFLogxK

U2 - https://doi.org/10.1007/978-3-662-44848-9_5

DO - https://doi.org/10.1007/978-3-662-44848-9_5

M3 - منشور من مؤتمر

SN - 9783662448472

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 66

EP - 81

BT - Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2014, Proceedings

T2 - European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2014

Y2 - 15 September 2014 through 19 September 2014

ER -

Avner O, Mannor S. Concurrent bandits and cognitive radio networks. In Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2014, Proceedings. PART 1 ed. 2014. p. 66-81. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). doi: https://doi.org/10.1007/978-3-662-44848-9_5

Concurrent bandits and cognitive radio networks

Abstract

Publication series

Conference

Keywords

All Science Journal Classification (ASJC) codes

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this