Reinforcement learning approaches and evaluation criteria for opportunistic spectrum access

Clement Robert; Christophe Moy; Cheng Xiang Wang

doi:10.1109/ICC.2014.6883535

Reinforcement learning approaches and evaluation criteria for opportunistic spectrum access

Clement Robert, Christophe Moy, Cheng Xiang Wang

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

11 Citations (Scopus)

Abstract

This paper deals with the learning and decision making issue for cognitive radio (CR). Two reinforcement-learning algorithms proposed in the literature are compared for opportunistic spectrum access (OSA): Upper Confidence Bound (UCB) algorithm and Weight Driven (WD) algorithm. This paper also introduces two new metrics in order to evaluate the machine learning algorithm performance for CR: effective cumulative regret and percentage of successful trials. They provide a fair evaluation means for CR performance.

Original language	English
Title of host publication	2014 IEEE International Conference on Communications, ICC 2014
Publisher	IEEE
Pages	1508-1513
Number of pages	6
ISBN (Print)	9781479920037
DOIs	https://doi.org/10.1109/ICC.2014.6883535
Publication status	Published - 2014
Event	IEEE International Conference on Communications 2014 - Sydney, NSW, Australia Duration: 10 Jun 2014 → 14 Jun 2014

Conference

Conference	IEEE International Conference on Communications 2014
Abbreviated title	ICC 2014
Country/Territory	Australia
City	Sydney, NSW
Period	10/06/14 → 14/06/14

Keywords

Cognitive radio
MAB
machine learning
opportunistic spectrum access
UCB

ASJC Scopus subject areas

Computer Networks and Communications

Access to Document

10.1109/ICC.2014.6883535

Cite this

@inproceedings{9af11d194c4347058ecb531e0fc919d6,

title = "Reinforcement learning approaches and evaluation criteria for opportunistic spectrum access",

abstract = "This paper deals with the learning and decision making issue for cognitive radio (CR). Two reinforcement-learning algorithms proposed in the literature are compared for opportunistic spectrum access (OSA): Upper Confidence Bound (UCB) algorithm and Weight Driven (WD) algorithm. This paper also introduces two new metrics in order to evaluate the machine learning algorithm performance for CR: effective cumulative regret and percentage of successful trials. They provide a fair evaluation means for CR performance.",

keywords = "Cognitive radio, MAB, machine learning, opportunistic spectrum access, UCB",

author = "Clement Robert and Christophe Moy and Wang, {Cheng Xiang}",

year = "2014",

doi = "10.1109/ICC.2014.6883535",

language = "English",

isbn = "9781479920037",

pages = "1508--1513",

booktitle = "2014 IEEE International Conference on Communications, ICC 2014",

publisher = "IEEE",

address = "United States",

note = "IEEE International Conference on Communications 2014, ICC 2014 ; Conference date: 10-06-2014 Through 14-06-2014",

}

TY - GEN

T1 - Reinforcement learning approaches and evaluation criteria for opportunistic spectrum access

AU - Robert, Clement

AU - Moy, Christophe

AU - Wang, Cheng Xiang

PY - 2014

Y1 - 2014

N2 - This paper deals with the learning and decision making issue for cognitive radio (CR). Two reinforcement-learning algorithms proposed in the literature are compared for opportunistic spectrum access (OSA): Upper Confidence Bound (UCB) algorithm and Weight Driven (WD) algorithm. This paper also introduces two new metrics in order to evaluate the machine learning algorithm performance for CR: effective cumulative regret and percentage of successful trials. They provide a fair evaluation means for CR performance.

AB - This paper deals with the learning and decision making issue for cognitive radio (CR). Two reinforcement-learning algorithms proposed in the literature are compared for opportunistic spectrum access (OSA): Upper Confidence Bound (UCB) algorithm and Weight Driven (WD) algorithm. This paper also introduces two new metrics in order to evaluate the machine learning algorithm performance for CR: effective cumulative regret and percentage of successful trials. They provide a fair evaluation means for CR performance.

KW - Cognitive radio

KW - MAB

KW - machine learning

KW - opportunistic spectrum access

KW - UCB

UR - http://www.scopus.com/inward/record.url?scp=84906996368&partnerID=8YFLogxK

U2 - 10.1109/ICC.2014.6883535

DO - 10.1109/ICC.2014.6883535

M3 - Conference contribution

AN - SCOPUS:84906996368

SN - 9781479920037

SP - 1508

EP - 1513

BT - 2014 IEEE International Conference on Communications, ICC 2014

PB - IEEE

T2 - IEEE International Conference on Communications 2014

Y2 - 10 June 2014 through 14 June 2014

ER -

Reinforcement learning approaches and evaluation criteria for opportunistic spectrum access

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this