Blockchained on-device federated learning

Hyesung Kim; Jihong Park; Mehdi Bennis; Seong Lyun Kim

doi:10.1109/LCOMM.2019.2921755

Blockchained on-device federated learning

Hyesung Kim, Jihong Park, Mehdi Bennis, Seong Lyun Kim

Department of Electrical and Electronic Engineering

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

582 Citations (Scopus)

Abstract

By leveraging blockchain, this letter proposes a blockchained federated learning (BlockFL) architecture where local learning model updates are exchanged and verified. This enables on-device machine learning without any centralized training data or coordination by utilizing a consensus mechanism in blockchain. Moreover, we analyze an end-to-end latency model of BlockFL and characterize the optimal block generation rate by considering communication, computation, and consensus delays.

Original language	English
Article number	8733825
Pages (from-to)	1279-1283
Number of pages	5
Journal	IEEE Communications Letters
Volume	24
Issue number	6
DOIs	https://doi.org/10.1109/LCOMM.2019.2921755
Publication status	Published - 2020 Jun

Bibliographical note

Publisher Copyright:
© 1997-2012 IEEE.

All Science Journal Classification (ASJC) codes

Modelling and Simulation
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/LCOMM.2019.2921755

Cite this

@article{54de89f62dc24789873c2a262717ea9b,

title = "Blockchained on-device federated learning",

abstract = "By leveraging blockchain, this letter proposes a blockchained federated learning (BlockFL) architecture where local learning model updates are exchanged and verified. This enables on-device machine learning without any centralized training data or coordination by utilizing a consensus mechanism in blockchain. Moreover, we analyze an end-to-end latency model of BlockFL and characterize the optimal block generation rate by considering communication, computation, and consensus delays.",

keywords = "On-device machine learning, blockchain, federated learning, latency",

author = "Hyesung Kim and Jihong Park and Mehdi Bennis and Kim, {Seong Lyun}",

note = "Publisher Copyright: {\textcopyright} 1997-2012 IEEE.",

year = "2020",

month = jun,

doi = "10.1109/LCOMM.2019.2921755",

language = "English",

volume = "24",

pages = "1279--1283",

journal = "IEEE Communications Letters",

issn = "1089-7798",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "6",

}

TY - JOUR

T1 - Blockchained on-device federated learning

AU - Kim, Hyesung

AU - Park, Jihong

AU - Bennis, Mehdi

AU - Kim, Seong Lyun

PY - 2020/6

Y1 - 2020/6

N2 - By leveraging blockchain, this letter proposes a blockchained federated learning (BlockFL) architecture where local learning model updates are exchanged and verified. This enables on-device machine learning without any centralized training data or coordination by utilizing a consensus mechanism in blockchain. Moreover, we analyze an end-to-end latency model of BlockFL and characterize the optimal block generation rate by considering communication, computation, and consensus delays.

AB - By leveraging blockchain, this letter proposes a blockchained federated learning (BlockFL) architecture where local learning model updates are exchanged and verified. This enables on-device machine learning without any centralized training data or coordination by utilizing a consensus mechanism in blockchain. Moreover, we analyze an end-to-end latency model of BlockFL and characterize the optimal block generation rate by considering communication, computation, and consensus delays.

KW - On-device machine learning

KW - blockchain

KW - federated learning

KW - latency

UR - http://www.scopus.com/inward/record.url?scp=85086451854&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85086451854&partnerID=8YFLogxK

U2 - 10.1109/LCOMM.2019.2921755

DO - 10.1109/LCOMM.2019.2921755

M3 - Article

AN - SCOPUS:85086451854

SN - 1089-7798

VL - 24

SP - 1279

EP - 1283

JO - IEEE Communications Letters

JF - IEEE Communications Letters

IS - 6

M1 - 8733825

ER -

Blockchained on-device federated learning

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this