Machine Learning Enabled Edge Computing: A Survey and Research Challenges

Abstract

AbstractEdge Computing has been regarded as a significant technology in 5G/B5G communication networks, which can improve the performance of network by deploying storage as well as computing resources on the edge of the network. At the same time, machine learning as a significant optimization approach has attracted wide attention in industry and academia. Hence, it is a natural trend to use machine learning to optimize the performance of edge computing. In addition, many excellent works on machine learning and edge computing have been done. Hence, it is necessary to survey these works. In this article, we survey the latest development for machine learning enabled edge computing. We first introduce the edge computing and machine learning separately and present the motivation for machine learning enabled edge computing. And then the research issues of the machine learning enabled edge computing from the perspective of user side and network side are presented. Finally, the research challenges and future directions are presented.KeywordsEdge computingMachine learningOptimizationResearch challengesSurvey