Online learning for "thing-adaptive" Fog Computing in IoT

Tianyi Chen; Yanning Shen; Qing Ling; Georgios B Giannakis

doi:10.1109/ACSSC.2017.8335425

Online learning for "thing-adaptive" Fog Computing in IoT

Tianyi Chen, Yanning Shen, Qing Ling, Georgios B Giannakis

Electrical and Computer Engineering

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

9 Scopus citations

Abstract

The present paper deals with online convex optimization involving time-varying loss functions and time-varying constraints. The constraints are revealed after making decisions, and allow instantaneous violations yet they must be satisfied in the long term. This setting fits nicely emerging online tasks such as fog computing, where online decisions need to flexibly adapt to the temporally unpredictable availability of resources. Tailored for heterogeneous systems such as those involved in the Internet of Things (IoT), a 'thing-adaptive' online saddle-point (TAOSP) scheme is developed, which automatically adjusts the stepsize to offer desirable task-specific learning rates. Performance here is assessed by: i) dynamic regret that generalizes the widely used static regret; and, ii) dynamic fit that captures the accumulated amount of constraint violations. Specifically, TAOSP is proved to simultaneously yield sub-linear dynamic regret and fit, provided that the best dynamic solutions vary slowly over time. Numerical tests in fog offloading tasks corroborate that our TAOSP approach outperforms the state of the art in speeding up computations.

Original language	English (US)
Title of host publication	Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017
Editors	Michael B. Matthews
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	664-668
Number of pages	5
ISBN (Electronic)	9781538618233
DOIs	https://doi.org/10.1109/ACSSC.2017.8335425
State	Published - Jul 2 2017
Event	51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017 - Pacific Grove, United States Duration: Oct 29 2017 → Nov 1 2017

Publication series

Name	Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017
Volume	2017-October

Other

Other	51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017
Country/Territory	United States
City	Pacific Grove
Period	10/29/17 → 11/1/17

Bibliographical note

Funding Information:
Work in this paper was supported by NSF 1509040, 1508993, 1509005, NSF China 61573331, NSF Anhui 1608085QF130, and CAS-XDA06011203.

Publisher Copyright:
© 2017 IEEE.

Keywords

Internet of Things
Online learning
convex optimization
mobile edge computing
saddle-point method

Publisher link

10.1109/ACSSC.2017.8335425

Find It

Find It at U of M Libraries

Cite this

Chen, T., Shen, Y., Ling, Q., & Giannakis, G. B. (2017). Online learning for "thing-adaptive" Fog Computing in IoT. In M. B. Matthews (Ed.), Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017 (pp. 664-668). (Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017; Vol. 2017-October). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ACSSC.2017.8335425

Online learning for "thing-adaptive" Fog Computing in IoT. / Chen, Tianyi; Shen, Yanning; Ling, Qing et al.
Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017. ed. / Michael B. Matthews. Institute of Electrical and Electronics Engineers Inc., 2017. p. 664-668 (Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017; Vol. 2017-October).

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

Chen, T, Shen, Y, Ling, Q & Giannakis, GB 2017, Online learning for "thing-adaptive" Fog Computing in IoT. in MB Matthews (ed.), Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017. Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017, vol. 2017-October, Institute of Electrical and Electronics Engineers Inc., pp. 664-668, 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017, Pacific Grove, United States, 10/29/17. https://doi.org/10.1109/ACSSC.2017.8335425

Chen T, Shen Y, Ling Q, Giannakis GB. Online learning for "thing-adaptive" Fog Computing in IoT. In Matthews MB, editor, Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 664-668. (Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017). doi: 10.1109/ACSSC.2017.8335425

Chen, Tianyi ; Shen, Yanning ; Ling, Qing et al. / Online learning for "thing-adaptive" Fog Computing in IoT. Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017. editor / Michael B. Matthews. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 664-668 (Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017).

@inproceedings{76f9af6a8af34a45a0155f26445fa391,

title = "Online learning for {"}thing-adaptive{"} Fog Computing in IoT",

abstract = "The present paper deals with online convex optimization involving time-varying loss functions and time-varying constraints. The constraints are revealed after making decisions, and allow instantaneous violations yet they must be satisfied in the long term. This setting fits nicely emerging online tasks such as fog computing, where online decisions need to flexibly adapt to the temporally unpredictable availability of resources. Tailored for heterogeneous systems such as those involved in the Internet of Things (IoT), a 'thing-adaptive' online saddle-point (TAOSP) scheme is developed, which automatically adjusts the stepsize to offer desirable task-specific learning rates. Performance here is assessed by: i) dynamic regret that generalizes the widely used static regret; and, ii) dynamic fit that captures the accumulated amount of constraint violations. Specifically, TAOSP is proved to simultaneously yield sub-linear dynamic regret and fit, provided that the best dynamic solutions vary slowly over time. Numerical tests in fog offloading tasks corroborate that our TAOSP approach outperforms the state of the art in speeding up computations.",

keywords = "Internet of Things, Online learning, convex optimization, mobile edge computing, saddle-point method",

author = "Tianyi Chen and Yanning Shen and Qing Ling and Giannakis, {Georgios B}",

note = "Funding Information: Work in this paper was supported by NSF 1509040, 1508993, 1509005, NSF China 61573331, NSF Anhui 1608085QF130, and CAS-XDA06011203. Publisher Copyright: {\textcopyright} 2017 IEEE.; 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017 ; Conference date: 29-10-2017 Through 01-11-2017",

year = "2017",

month = jul,

day = "2",

doi = "10.1109/ACSSC.2017.8335425",

language = "English (US)",

series = "Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017",

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

pages = "664--668",

editor = "Matthews, {Michael B.}",

booktitle = "Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017",

}

TY - GEN

T1 - Online learning for "thing-adaptive" Fog Computing in IoT

AU - Chen, Tianyi

AU - Shen, Yanning

AU - Ling, Qing

AU - Giannakis, Georgios B

PY - 2017/7/2

Y1 - 2017/7/2

N2 - The present paper deals with online convex optimization involving time-varying loss functions and time-varying constraints. The constraints are revealed after making decisions, and allow instantaneous violations yet they must be satisfied in the long term. This setting fits nicely emerging online tasks such as fog computing, where online decisions need to flexibly adapt to the temporally unpredictable availability of resources. Tailored for heterogeneous systems such as those involved in the Internet of Things (IoT), a 'thing-adaptive' online saddle-point (TAOSP) scheme is developed, which automatically adjusts the stepsize to offer desirable task-specific learning rates. Performance here is assessed by: i) dynamic regret that generalizes the widely used static regret; and, ii) dynamic fit that captures the accumulated amount of constraint violations. Specifically, TAOSP is proved to simultaneously yield sub-linear dynamic regret and fit, provided that the best dynamic solutions vary slowly over time. Numerical tests in fog offloading tasks corroborate that our TAOSP approach outperforms the state of the art in speeding up computations.

AB - The present paper deals with online convex optimization involving time-varying loss functions and time-varying constraints. The constraints are revealed after making decisions, and allow instantaneous violations yet they must be satisfied in the long term. This setting fits nicely emerging online tasks such as fog computing, where online decisions need to flexibly adapt to the temporally unpredictable availability of resources. Tailored for heterogeneous systems such as those involved in the Internet of Things (IoT), a 'thing-adaptive' online saddle-point (TAOSP) scheme is developed, which automatically adjusts the stepsize to offer desirable task-specific learning rates. Performance here is assessed by: i) dynamic regret that generalizes the widely used static regret; and, ii) dynamic fit that captures the accumulated amount of constraint violations. Specifically, TAOSP is proved to simultaneously yield sub-linear dynamic regret and fit, provided that the best dynamic solutions vary slowly over time. Numerical tests in fog offloading tasks corroborate that our TAOSP approach outperforms the state of the art in speeding up computations.

KW - Internet of Things

KW - Online learning

KW - convex optimization

KW - mobile edge computing

KW - saddle-point method

UR - http://www.scopus.com/inward/record.url?scp=85050945244&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050945244&partnerID=8YFLogxK

U2 - 10.1109/ACSSC.2017.8335425

DO - 10.1109/ACSSC.2017.8335425

M3 - Conference contribution

AN - SCOPUS:85050945244

T3 - Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017

SP - 664

EP - 668

BT - Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017

A2 - Matthews, Michael B.

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017

Y2 - 29 October 2017 through 1 November 2017

ER -

Online learning for "thing-adaptive" Fog Computing in IoT

Abstract

Publication series

Other

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this