King's College London

Research portal

Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Standard

Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty. / Sun, Zhenfeng; Nakhai, Mohammad.

2020 IEEE International Conference on Communications, ICC 2020 - Proceedings. Institute of Electrical and Electronics Engineers ( IEEE ), 2020. 9149327 (IEEE International Conference on Communications; Vol. 2020-June).

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Harvard

Sun, Z & Nakhai, M 2020, Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty. in 2020 IEEE International Conference on Communications, ICC 2020 - Proceedings., 9149327, IEEE International Conference on Communications, vol. 2020-June, Institute of Electrical and Electronics Engineers ( IEEE ). https://doi.org/10.1109/ICC40277.2020.9149327

APA

Sun, Z., & Nakhai, M. (2020). Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty. In 2020 IEEE International Conference on Communications, ICC 2020 - Proceedings [9149327] (IEEE International Conference on Communications; Vol. 2020-June). Institute of Electrical and Electronics Engineers ( IEEE ). https://doi.org/10.1109/ICC40277.2020.9149327

Vancouver

Sun Z, Nakhai M. Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty. In 2020 IEEE International Conference on Communications, ICC 2020 - Proceedings. Institute of Electrical and Electronics Engineers ( IEEE ). 2020. 9149327. (IEEE International Conference on Communications). https://doi.org/10.1109/ICC40277.2020.9149327

Author

Sun, Zhenfeng ; Nakhai, Mohammad. / Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty. 2020 IEEE International Conference on Communications, ICC 2020 - Proceedings. Institute of Electrical and Electronics Engineers ( IEEE ), 2020. (IEEE International Conference on Communications).

Bibtex Download

@inproceedings{cd2e9a1a0d904961890a871763aea9a0,
title = "Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty",
abstract = "This work deals with the task offloading problem for multiple cellular edge devices in a multi-access edge computing (MEC) infrastructure attached to a base-station (BS). In order to minimize the overall task computing-communication delay through coping with time-varying cost and constraint functions with unknown statistics on-the-go, we propose a novel distributed bandit optimization (DBO) algorithm which runs based on the projected dual gradient iterations and a single broadcast communicating the MEC states to the SDs at the end of each time-slot. To track the performance of the proposed online learning algorithm over time, we define a dynamic regret to assess the closeness of the underlying delay cost of the DBO to a clairvoyant dynamic optimum and an aggregate violation metric to evaluate the asymptotic satisfaction of the constraints. We derive lower and upper bounds for dynamic regret as well as an upper-bound for the aggregate violation and show that the upper- bounds are sub-linear under sub-linear accumulated hindsight variations. The simulation results and comparisons confirm the effectiveness of the proposed algorithm in the long run.",
keywords = "Online learning, aggregate violation, dynamic regret, multi-access edge computing, projected dual gradient",
author = "Zhenfeng Sun and Mohammad Nakhai",
year = "2020",
month = jun,
doi = "10.1109/ICC40277.2020.9149327",
language = "English",
series = "IEEE International Conference on Communications",
publisher = "Institute of Electrical and Electronics Engineers ( IEEE )",
booktitle = "2020 IEEE International Conference on Communications, ICC 2020 - Proceedings",

}

RIS (suitable for import to EndNote) Download

TY - GEN

T1 - Edge Intelligence: Distributed Task Offloading and Service Management under Uncertainty

AU - Sun, Zhenfeng

AU - Nakhai, Mohammad

PY - 2020/6

Y1 - 2020/6

N2 - This work deals with the task offloading problem for multiple cellular edge devices in a multi-access edge computing (MEC) infrastructure attached to a base-station (BS). In order to minimize the overall task computing-communication delay through coping with time-varying cost and constraint functions with unknown statistics on-the-go, we propose a novel distributed bandit optimization (DBO) algorithm which runs based on the projected dual gradient iterations and a single broadcast communicating the MEC states to the SDs at the end of each time-slot. To track the performance of the proposed online learning algorithm over time, we define a dynamic regret to assess the closeness of the underlying delay cost of the DBO to a clairvoyant dynamic optimum and an aggregate violation metric to evaluate the asymptotic satisfaction of the constraints. We derive lower and upper bounds for dynamic regret as well as an upper-bound for the aggregate violation and show that the upper- bounds are sub-linear under sub-linear accumulated hindsight variations. The simulation results and comparisons confirm the effectiveness of the proposed algorithm in the long run.

AB - This work deals with the task offloading problem for multiple cellular edge devices in a multi-access edge computing (MEC) infrastructure attached to a base-station (BS). In order to minimize the overall task computing-communication delay through coping with time-varying cost and constraint functions with unknown statistics on-the-go, we propose a novel distributed bandit optimization (DBO) algorithm which runs based on the projected dual gradient iterations and a single broadcast communicating the MEC states to the SDs at the end of each time-slot. To track the performance of the proposed online learning algorithm over time, we define a dynamic regret to assess the closeness of the underlying delay cost of the DBO to a clairvoyant dynamic optimum and an aggregate violation metric to evaluate the asymptotic satisfaction of the constraints. We derive lower and upper bounds for dynamic regret as well as an upper-bound for the aggregate violation and show that the upper- bounds are sub-linear under sub-linear accumulated hindsight variations. The simulation results and comparisons confirm the effectiveness of the proposed algorithm in the long run.

KW - Online learning

KW - aggregate violation

KW - dynamic regret

KW - multi-access edge computing

KW - projected dual gradient

UR - http://www.scopus.com/inward/record.url?scp=85089416066&partnerID=8YFLogxK

U2 - 10.1109/ICC40277.2020.9149327

DO - 10.1109/ICC40277.2020.9149327

M3 - Conference contribution

T3 - IEEE International Conference on Communications

BT - 2020 IEEE International Conference on Communications, ICC 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers ( IEEE )

ER -

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454