题目：Federated Learning in Vehicular Networks:A Brief Introduction

报告人：余荣教授 广东工业大学

时间：2019年11月6日   14：30

地点：理六栋220

邀请人：张涵

报告内容摘要：Federated learning is a newly emerged distributed machine learning paradigm,which aims to efficiently exploit distributed data from wireless clients for deep learning model training, and meanwhile, effectively protect individual privacy.The study on federated learning in vehicular networks gives rise to a new frontier for the interdiscipline of wireless networking and machine learning, which is promising to meet the ever-increasing demands of Artificial Intelligence (AI) applications in Intelligent Connected Vehicles (ICV). In this talk, the state-ofthe-art of federated learning will be firstly overviewed. After that, the potential of federated learning in vehicular networks will be comprehensively presented.An exemplary case of selective model aggregation will be discussed to illustrate the underlying challenges, solving approaches, and open issues of federated learning in vehicular networks. In addition, an on-board prototype for demonstration will be exhibited with its typical use cases to intuitionally show the future application in ICV and Intelligent Transportation System (ITS).

个人简介：Rong Yu received his Ph.D. degree from Department of Electronic Engineering at Tsinghua University, China, in 2007. After that, he worked in the School of Electronic and Information Engineering at South China University of Technology. In 2010, he joined the School of Automation at Guangdong University of Technology, where he is now a full professor. His research interests include wireless networking and mobile computing such as Edge Intelligence, Connected Vehicles, Smart Grid, and Internet of Things. He is the co-inventor of over 40 patents and author or co-author of over 100 international journal and conference papers. He was the member of home networking standard committee in China, where he led the standardization work of three standards.