[ "周知(1993- ),男,中山大学数据科学与计算机学院特聘研究员,主要研究方向为云计算、边缘计算和分布式系统。" ]
[ "于帅(1986- ),男,博士,中山大学数据科学与计算机学院在站博士后,主要研究方向为无线通信、移动计算、机器学习等。" ]
[ "陈旭(1986- ),男,中山大学数据科学与计算机学院教授、博士生导师,主要研究方向为边缘计算、边缘智能、智能物联网等。" ]
网络首发:2019-03,
纸质出版:2019-03-15
移动端阅览
周知, 于帅, 陈旭. 边缘智能:边缘计算与人工智能融合的新范式[J]. 大数据, 2019,5(2):2019013-1.
Zhi ZHOU, Shuai YU, Xu CHEN. Edge intelligence:a new nexus of edge computing and artificial intelligence[J]. Big Data Research, 2019, 5(2): 2019013-1.
周知, 于帅, 陈旭. 边缘智能:边缘计算与人工智能融合的新范式[J]. 大数据, 2019,5(2):2019013-1. DOI: 10.11959/j.issn.2096-0271.2019013.
Zhi ZHOU, Shuai YU, Xu CHEN. Edge intelligence:a new nexus of edge computing and artificial intelligence[J]. Big Data Research, 2019, 5(2): 2019013-1. DOI: 10.11959/j.issn.2096-0271.2019013.
边缘计算与人工智能这两种高速发展的新技术之间存在着彼此赋能的巨大潜力。通过3个研究案例,展示协同边缘计算和人工智能这两种技术如何促进各自的进一步发展。从边缘计算赋能人工智能的维度,针对深度学习模型在网络边缘侧的部署,提出了基于边端协同的深度学习按需加速框架,通过协同优化模型分割和模型精简策略,实现时延约束下的高精度模型推理。从人工智能赋能边缘计算的维度,针对边缘计算服务的放置问题,提出了基于在线学习的自适应边缘服务放置机制和基于因子图模型的预测性边缘服务迁移方法。
Artificial intelligence (AI) and edge computing (EC) represent two of today’s most popular technologies.There is a great potential to coordinate these two emerging techniques to facilitate the further advent of both sides.Through three research cases
the profound benefits were demonstrated when AI and EC synergize.Specifically
from the perspective of EC for AI
to efficiently run deep learning at the network edge
a collaborative and on-demand deep neural network (DNN) co-inference framework with device-edge synergy was proposed.By applying DNN partitioning and right-sizing
it minimizes the inference latency under target accuracy.On the other hand
from the perspective of AI for EC
for the dynamical placement of edge computing services
two methods were proposed:an online-learning based adaptive service migration strategy and a factor graph model driven predictive service migration technique.
LECUN Y , BENGIO Y , HINTON G . Deep learning [J ] . Nature , 2015 , 521 ( 7553 ):436.
SHI W S , CAO J , ZHANG Q Y , et al . Edge computing:vision and challenges [J ] . IEEE Internet of Things Journal , 2016 , 3 ( 5 ): 637 - 646 .
IANDOLA F N , MOSKEWICZ M W , ASHRAF K , et al . SqueezeNet:AlexNetlevel accuracy with 50x fewer parameters and<0.5MB model size [J ] . Computer Science , 2016 ,arXiv:1602.07360.
WU J X , CONG L , WANG Y H , et al . Quantized convolutional neural networks for mobile devices [C ] // IEEE Conference on Computer Vision and Pattern Recognition (CVPR),June 26-July 1,2016,Las Vegas,USA . Piscataway:IEEE Press , 2016 : 4820 - 4828 .
HAN S , MAO H Z , DALLY W J . Deep compression:compressing deep neural networks with pruning,trained quantization and Huffman coding [J ] . Computer Science , 2015 ,arXiv:1510.00149.
KANG Y P , HAUSWALD J , GAO C , et al . Neurosurgeon:collaborative intelligence between the cloud and mobile edge [C ] // The 22nd International Conference on Architectural Support for Programming Languages and Operating Systems,April l8-12,2017,Xi’an,China . New York:ACM Press , 2017 : 615 - 629 .
LANE N , BHATTACHARYA S , MATHUR A , et al . DXTK:enabling resource-efficient deep learning on mobile and embedded devices with the DeepX toolkit [C ] // The 8th EAI International Conference on Mobile Computing,Applications and Services,November 30-December 1,2016,Cambridge,UK . New York:ACM Press , 2017 : 98 - 107 .
TEERAPITTAYANON S , MCDANEL B , KUNG H T . BranchyNet:fast inference via early exiting from deep neural networks [J ] . Computer Science , 2017 ,arXiv:1709.01686.
LI E , ZHOU Z , CHEN X . Edge intelligence:on-demand deep learning model co-inference with device-edge synergy [C ] // ACM Workshop on Mobile Edge Communications,August 20,2018,Budapest,Hungary . New York:ACM Press , 2018 : 31 - 36 .
MACH P AND BECVAR Z . Mobile edge computing:a survey on architecture and computation offloading [J ] . IEEE Communications Surveys & Tutorials , 2017 , 3 ( 19 ): 1628 - 1656 .
WANG S , XU J , ZHANG N , et al . A survey on service migration in mobile edge computing [J ] . IEEE Access , 2018 , 6 ( 24 ): 511 - 528 .
OUYANG T , LI R , CHEN X , et al . Adaptive user-managed service placement for mobile edge computing:an online learning approach [C ] // IEEE International Conference on Computer Communications,April 28-May 2,2019,Paris,France . Piscataway:IEEE Press , 2019 .
AGRAWAL S , GOYAL N . Thompson sampling for contextual bandits with linear payoffs [C ] // The 30th International Conference on International Conference on Machine Learning,June 16-21,2013,Atlanta,USA.[S.l.:s.n . ] , 2013 : 1220 - 1228 .
RUSSO D J , ROY B V , KAZEROUNI A , et al . A tutorial on thompson sampling [J ] . Foundations and Trends in Machine Learning , 2018 , 1 ( 11 ): 1 - 96 .
WU Q , CHEN X , ZHOU Z , et al . Mobile social data learning for user-centric location prediction with application in mobile edge service migration [J ] . IEEE Internet of Things Journal , 2019 :accepted. 1 - 96 .
KSCHISCHANG F R , FREY B J , LOELIGER H A . Factor graphs and the sum-product algorithm [J ] . IEEE Transactions on Information Theory , 2002 , 2 ( 47 ): 498 - 519 .
0
浏览量
2060
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621