报告题目:Trustworthy Distributed Traffic Prediction through Graph Neural Network 基于图神经网络的可信分布式交通预测
报告摘要:
Due to the rapid development of Internet of Things (IoT) technology, GPS modules have been widely used throughout various kinds of mobile devices. These mobile devices collect massive vehicle trajectory and empower traffic prediction-based tasks of the intelligent transportation system (ITS), including travel time estimation (TTE), route recovery and road condition analysis. Most previous works model both road network and vehicle trajectory by learning their spatio-temporal characteristics to conduct traffic prediction. Among them, graph neural network (GNN) has become a popular option for graph-structured traffic data. However, there are still two key issues for trustworthy distributed traffic prediction in real-world location-based services, i.e., data sparsity and privacy protection. In this report, I will introduce my corresponding research progress about how to solve the above issues through GNN.
随着物联网技术的快速发展,各类移动设备的GPS模块收集了海量的车辆轨迹数据,这些时空大数据为智能交通系统的发展提供了有力支撑,旅行时间估计、轨迹路径恢复和道路状况分析等交通预测任务成为了该系统的研究热点。在相关的交通预测研究工作中,道路网络和车辆轨迹的时空属性使得图神经网络成为了一个流行的方法选择。尽管时空图神经网络在交通预测任务上取得了很高的预测精度,但从实际的位置服务应用角度出发,基于图神经网络的可信分布式交通预测仍存在两个关键问题,即通信带宽导致的数据稀疏问题和用户上传敏感轨迹的隐私保护问题。在本报告中,主讲人将介绍其在上述问题的研究进展。
个人简介:
张志文,先后在南开大学获得人工智能学士学位和控制科学硕士学位,2022年取得日本东京大学社会文化环境专业的博士学位。目前是日本LocationMind公司的数据科学家,研究方向集中在城市计算和时空数据分析,其相关研究成果发表在T-ITS,TKDE和IoTJ等学术期刊上。
主办单位:澳门人威尼斯官方
报告时间:2023年4月21日(星期五)上午9:00
会议室地址:腾讯会议489-924-851