报告题目：面向时空交通数据修复的拉普拉斯卷积模型 (Laplacian convolutional representation for traffic time series imputation)
报告时间：2024年07月11日上午10:00-11:30
讲座地点：经管学院B309
报告人：陈新宇 博士后
邀请人：贺冬冬 助理教授
报告内容和摘要：
Spatiotemporal traffic data imputation is of great significance in intelligent transportation systems and data-driven decision-making processes. To perform efficient learning and accurate reconstruction from partially observed traffic data, we assert the importance of characterizing both global and local trends in time series. In this study, we first introduce a Laplacian kernel to temporal regularization for characterizing local trends in traffic time series, which can be formulated as a circular convolution. Then, we develop a low-rank Laplacian convolutional representation (LCR) model by putting the circulant matrix nuclear norm and the Laplacian kernelized temporal regularization together, which is proved to meet a unified framework that has a fast Fourier transform solution in log-linear time complexity. Through extensive experiments on several traffic datasets, we demonstrate the superiority of LCR over several baseline models for imputing traffic time series of various time series behaviors and reconstructing sparse speed fields of vehicular traffic flow. The proposed LCR model is also an efficient solution to large-scale traffic data imputation over the existing imputation models.
报告人简介：
陈新宇博士，目前在美国麻省理工学院从事博士后研究，此前在加拿大蒙特利尔大学获得博士学位，研究方向为机器学习、数据科学、城市科学、智能交通，研究兴趣包括机器学习理论与算法、时空数据建模技术、张量计算与数据挖掘。以第一作者在计算机领域顶刊 IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) 与 IEEE Transactions on Knowledge and Data Engineering (TKDE) 发表学术论文 3 篇，在智能交通领域顶刊 Transportation Research Part C: Emerging Technologies 与 IEEE Transactions on Intelligent Transportation Systems 发表学术论文 6 篇，所有论文共计被引用超过 1100 次，其中，2 篇论文曾入选 ESI 热点论文、 2 篇论文被收录为 ESI 高被引论文。
上一条：基于区块链的数据要素市场关键技术与应用
下一条：Impact of channel co-opetition and taxes on a multinational firm’s resilient local production decision
【关闭】
