西安市移动边缘计算及安全重点实验室
首页 | 实验室概况 | 实验室管理 | 新闻公告 | 科学研究 | 研究队伍 | 开放交流 | 联系我们
通知公告
 新闻动态 
 通知公告 
当前位置: 首页 >> 新闻公告
Coded Stochastic ADMM for Decentralized Consensus Optimization with Edge Computing讲座通知
2021年09月06日   审核人:

讲座名称:Coded Stochastic ADMM for Decentralized Consensus Optimization with Edge Computing
讲座时间:2021-08-02 15: 00
讲座人:Ming Xiao
讲座地点:腾讯会议(ID: 405 245 124)
讲座人介绍:
Ming Xiao (Senior Member of IEEE) received Bachelor and Master degrees in Engineering from the University of Electronic Science and Technology of China. He received Ph.D degree from Chalmers University of technology, Sweden. From November 2007 to now, he has been, Royal Institute of Technology (KTH), Sweden, where he is currently an Associate Professor. He has been an Associate Editor for IEEE Transaction on Communications, IEEE Communications Letters, IEEE Transactions on Wireless Communciations. He also services as the lead guest editor for IEEE Journal on Selected Areas in Communications (JSAC) special issue on "Millimeter wave communications for future mobile networks" and Guest Editor for IEEE Internet of Things, Journals. He is listed by Stanford Research as world top 2% researchers in 2021.
讲座内容:
Big data, including applications with high security requirements, are often collected and stored on multiple heterogeneous devices, such as mobile devices, drones and vehicles. Due to the limitations of communication costs and security requirements, it is of paramount importance to analyze information in a decentralized way instead of transmitting data to a fusion center. To train large-scale machine learning models, edge/fog computing is often leveraged as an alternative to centralized learning. We consider the problem of learning model parameters in a multi-agent system with data locally processed via distributed edge nodes. A class of mini-batch stochastic alternating direction method of multipliers (ADMM) algorithms is explored to develop the distributed learning model. To address two main critical challenges in distributed learning systems, i.e., communication bottleneck and straggler nodes (nodes with slow responses), error-control-coding based stochastic incremental ADMM is investigated.  Given an appropriate mini-batch size, we show that the mini-batch stochastic ADMM based method converges in a rate of 𝑂(1∕√𝑘)where 𝑘 denotes the number of iterations. Through numerical experiments, it is revealed that the proposed algorithm is communication-efficient, rapidly responding and robust in the presence of straggler nodes compared with state of the art algorithms.
主办单位:ISN国家重点实验室|现代无线信息网络111基地|通信工程学院|西安市移动边缘计算及安全重点实验室

关闭窗口

版权所有 西安市移动边缘计算及安全重点实验室 技术支持:西安聚力
Xi'an Key Laboratory of Mobile Edge Computing and Security