Home > Research > Publications & Outputs > Distributed service caching with deep reinforce...

Research

Associated organisational unit

Text available via DOI:

View graph of relations

Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Forthcoming

Standard

Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI. / Liu, Wei; Bilal, Muhammad; Shi, Yuzhe et al.
In: Digital Communications and Networks, 17.11.2024.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Liu, W, Bilal, M, Shi, Y & Xu, X 2024, 'Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI', Digital Communications and Networks. https://doi.org/10.1016/j.dcan.2024.11.009

APA

Liu, W., Bilal, M., Shi, Y., & Xu, X. (in press). Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI. Digital Communications and Networks. https://doi.org/10.1016/j.dcan.2024.11.009

Vancouver

Liu W, Bilal M, Shi Y, Xu X. Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI. Digital Communications and Networks. 2024 Nov 17. doi: 10.1016/j.dcan.2024.11.009

Author

Liu, Wei ; Bilal, Muhammad ; Shi, Yuzhe et al. / Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI. In: Digital Communications and Networks. 2024.

Bibtex

@article{05cf7ddb0bc24f7492e78fd234e25f75,
title = "Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI",
abstract = "Increasing reliance on large-scale AI models has led to rising demand for intelligent services. The centralized cloud computing approach has limitations in terms of data transfer efficiency and response time, and as a result many service providers have begun to deploy edge servers to cache intelligent services in order to reduce transmission delay and communication energy consumption. However, finding the optimal service caching strategy remains a significant challenge due to the stochastic nature of service requests and the bulky nature of intelligent services. To deal with this we propose a distributed service caching scheme integrating deep reinforcement learning (DRL) with mobility prediction, which we refer to as DSDM. Specifically, we employ the D3QN (Deep Double Dueling Q-Network) framework to integrate Long Short-Term Memory (LSTM) predicted mobile device locations into the service caching replacement algorithm and adopt the distributed multi-agent approach for learning and training. Experimental results demonstrate that DSDM achieves significant performance improvements in reducing communication energy consumption compared to traditional methods across various scenarios.",
author = "Wei Liu and Muhammad Bilal and Yuzhe Shi and Xiaolong Xu",
year = "2024",
month = nov,
day = "17",
doi = "10.1016/j.dcan.2024.11.009",
language = "English",
journal = "Digital Communications and Networks",
issn = "2352-8648",
publisher = "Elsevier",

}

RIS

TY - JOUR

T1 - Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI

AU - Liu, Wei

AU - Bilal, Muhammad

AU - Shi, Yuzhe

AU - Xu, Xiaolong

PY - 2024/11/17

Y1 - 2024/11/17

N2 - Increasing reliance on large-scale AI models has led to rising demand for intelligent services. The centralized cloud computing approach has limitations in terms of data transfer efficiency and response time, and as a result many service providers have begun to deploy edge servers to cache intelligent services in order to reduce transmission delay and communication energy consumption. However, finding the optimal service caching strategy remains a significant challenge due to the stochastic nature of service requests and the bulky nature of intelligent services. To deal with this we propose a distributed service caching scheme integrating deep reinforcement learning (DRL) with mobility prediction, which we refer to as DSDM. Specifically, we employ the D3QN (Deep Double Dueling Q-Network) framework to integrate Long Short-Term Memory (LSTM) predicted mobile device locations into the service caching replacement algorithm and adopt the distributed multi-agent approach for learning and training. Experimental results demonstrate that DSDM achieves significant performance improvements in reducing communication energy consumption compared to traditional methods across various scenarios.

AB - Increasing reliance on large-scale AI models has led to rising demand for intelligent services. The centralized cloud computing approach has limitations in terms of data transfer efficiency and response time, and as a result many service providers have begun to deploy edge servers to cache intelligent services in order to reduce transmission delay and communication energy consumption. However, finding the optimal service caching strategy remains a significant challenge due to the stochastic nature of service requests and the bulky nature of intelligent services. To deal with this we propose a distributed service caching scheme integrating deep reinforcement learning (DRL) with mobility prediction, which we refer to as DSDM. Specifically, we employ the D3QN (Deep Double Dueling Q-Network) framework to integrate Long Short-Term Memory (LSTM) predicted mobile device locations into the service caching replacement algorithm and adopt the distributed multi-agent approach for learning and training. Experimental results demonstrate that DSDM achieves significant performance improvements in reducing communication energy consumption compared to traditional methods across various scenarios.

U2 - 10.1016/j.dcan.2024.11.009

DO - 10.1016/j.dcan.2024.11.009

M3 - Journal article

JO - Digital Communications and Networks

JF - Digital Communications and Networks

SN - 2352-8648

ER -