Asynchronous broadcast-based decentralized learning in sensor networks

Liang Zhao; Wen Zhan Song; Xiaojing Ye; Yujie Gu

doi:10.1080/17445760.2017.1294690

Asynchronous broadcast-based decentralized learning in sensor networks

Liang Zhao
, Wen Zhan Song
, Xiaojing Ye
, Yujie Gu

Research output: Contribution to journal › Article › peer-review

17 Scopus citations

Abstract

In this paper, we study the problem of decentralized learning in sensor networks in which local learners estimate and reach consensus to the quantity of interest inferred globally while communicating only with their immediate neighbours. The main challenge lies in reducing the communication cost in the network, which involves inter-node synchronisation and data exchange. To address this issue, a novel asynchronous broadcast-based decentralized learning algorithm is proposed. Furthermore, we prove that the iterates generated by the developed decentralized method converge to a consensual optimal solution (model). Numerical results demonstrate that it is a promising approach for decentralized learning in sensor networks.

Original language	English
Pages (from-to)	589-607
Number of pages	19
Journal	International Journal of Parallel, Emergent and Distributed Systems
Volume	33
Issue number	6
DOIs	https://doi.org/10.1080/17445760.2017.1294690
State	Published - 2 Nov 2018
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 2017, © 2017 Informa UK Limited, trading as Taylor & Francis Group.

Funding

This work was partially supported by National Science Foundation [grant number 1066391], [grant number 1442630], and [grant number 1125165].

Funders	Funder number
National Science Foundation	1066391, 1442630, 1125165, 1663709

Keywords

Asynchronous algorithm
big data
decentralized learning
sensor network

Access to Document

10.1080/17445760.2017.1294690

Cite this

@article{cc52b5d52b58478d9edbf1032e00caba,

title = "Asynchronous broadcast-based decentralized learning in sensor networks",

abstract = "In this paper, we study the problem of decentralized learning in sensor networks in which local learners estimate and reach consensus to the quantity of interest inferred globally while communicating only with their immediate neighbours. The main challenge lies in reducing the communication cost in the network, which involves inter-node synchronisation and data exchange. To address this issue, a novel asynchronous broadcast-based decentralized learning algorithm is proposed. Furthermore, we prove that the iterates generated by the developed decentralized method converge to a consensual optimal solution (model). Numerical results demonstrate that it is a promising approach for decentralized learning in sensor networks.",

keywords = "Asynchronous algorithm, big data, decentralized learning, sensor network",

author = "Liang Zhao and Song, \{Wen Zhan\} and Xiaojing Ye and Yujie Gu",

note = "Publisher Copyright: {\textcopyright} 2017, {\textcopyright} 2017 Informa UK Limited, trading as Taylor \& Francis Group.",

year = "2018",

month = nov,

day = "2",

doi = "10.1080/17445760.2017.1294690",

language = "אנגלית",

volume = "33",

pages = "589--607",

journal = "International Journal of Parallel, Emergent and Distributed Systems",

issn = "1744-5760",

publisher = "Taylor and Francis Ltd.",

number = "6",

}

TY - JOUR

T1 - Asynchronous broadcast-based decentralized learning in sensor networks

AU - Zhao, Liang

AU - Song, Wen Zhan

AU - Ye, Xiaojing

AU - Gu, Yujie

PY - 2018/11/2

Y1 - 2018/11/2

N2 - In this paper, we study the problem of decentralized learning in sensor networks in which local learners estimate and reach consensus to the quantity of interest inferred globally while communicating only with their immediate neighbours. The main challenge lies in reducing the communication cost in the network, which involves inter-node synchronisation and data exchange. To address this issue, a novel asynchronous broadcast-based decentralized learning algorithm is proposed. Furthermore, we prove that the iterates generated by the developed decentralized method converge to a consensual optimal solution (model). Numerical results demonstrate that it is a promising approach for decentralized learning in sensor networks.

AB - In this paper, we study the problem of decentralized learning in sensor networks in which local learners estimate and reach consensus to the quantity of interest inferred globally while communicating only with their immediate neighbours. The main challenge lies in reducing the communication cost in the network, which involves inter-node synchronisation and data exchange. To address this issue, a novel asynchronous broadcast-based decentralized learning algorithm is proposed. Furthermore, we prove that the iterates generated by the developed decentralized method converge to a consensual optimal solution (model). Numerical results demonstrate that it is a promising approach for decentralized learning in sensor networks.

KW - Asynchronous algorithm

KW - big data

KW - decentralized learning

KW - sensor network

UR - https://www.scopus.com/pages/publications/85014486701

U2 - 10.1080/17445760.2017.1294690

DO - 10.1080/17445760.2017.1294690

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:85014486701

SN - 1744-5760

VL - 33

SP - 589

EP - 607

JO - International Journal of Parallel, Emergent and Distributed Systems

JF - International Journal of Parallel, Emergent and Distributed Systems

IS - 6

ER -

Asynchronous broadcast-based decentralized learning in sensor networks

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this