《计算机应用研究》|Application Research of Computers

移动边缘计算中基于深度强化学习的计算卸载调度方法

Deep reinforcement learning based offloading scheduling in mobile edge computing

免费全文下载 (已被下载 次)  
获取PDF全文
作者 詹文翰,王瑾,朱清新,段翰聪,叶娅兰
机构 1.电子科技大学 a.信息与软件工程学院;b.计算机科学与工程学院,成都 611731;2.埃克塞特大学 计算机系,英国 埃克塞特 EX44RN
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2021)01-048-0241-05
DOI 10.19734/j.issn.1001-3695.2019.10.0594
摘要 针对移动边缘计算中具有依赖关系的任务的卸载决策问题,提出一种基于深度强化学习的任务卸载调度方法,以最小化应用程序的执行时间。任务调度的过程被描述为一个马尔可夫决策过程,其调度策略由所提出的序列到序列深度神经网络表示,并通过近端策略优化(proximal policy optimization)方法进行训练。仿真实验表明,所提出的算法具有良好的收敛能力,并且在不同环境下的表现均优于所对比的六个基线算法,证明了该方法的有效性和可靠性。
关键词 移动边缘计算; 计算卸载; 任务调度; 深度强化学习
基金项目 国家自然科学基金面上项目(61871096,61976047)
四川省科技厅重点研发项目(2019YFG0122)
本文URL http://www.arocmag.com/article/01-2021-01-048.html
英文标题 Deep reinforcement learning based offloading scheduling in mobile edge computing
作者英文名 Zhan Wenhan, Wang Jin, Zhu Qingxin, Duan Hancong, Ye Yalan
机构英文名 1.a.School of Information & Software Engineering,b.School of Computer Science & Engineering,University of Electronic Science & Technology of China,Chengdu 611731,China;2.College of Engineering,Mathematics & Physical Sciences,University of Exeter,Exeter EX44RN,UK
英文摘要 Aiming at the problem of task offloading with dependency in mobile edge computing, this paper proposed a deep reinforcement learning based offloading scheduling method to minimize the execution time of mobile applications. This method described the process of task scheduling as a Markov decision process. It adopted a sequence to sequence deep neural network to represent the scheduling policy, and then trained the deep neural network with the proximal policy optimization method. Simulation results show that the proposed method has good convergence ability and outperforms six baseline algorithms in different environments, demonstrating the effectiveness and reliability of the proposed method.
英文关键词 mobile edge computing; computation offloading; task scheduling; deep reinforcement learning
参考文献 查看稿件参考文献
 
收稿日期 2019/10/13
修回日期 2019/11/29
页码 241-245,263
中图分类号 TP391
文献标志码 A