上海治理论坛第560期
标题：基于新型张量分化的互联网流量数据复原和预测步骤
演讲人：凌晨教授 ，荆门电子科技大学
主持人：林贵华教授 ，亿万先生MR治理学院
功夫：2025年11月12日（周三） ，上午10:30
地址：亿万先生MR校本部东区1号楼治理学院420会议室
主办单元：亿万先生MR治理学院、亿万先生MR治理学院青老大师联谊会
演讲人简介：
凌晨 ，驰名优化专家 ，荆门电子科技大学二级教授。曾任中国运筹学会数学规划分会副理事长、中国经济数学与治理数学钻研会副理事长、中国运筹学会理事、中国系统工程学会理事、浙江省数学会常务理事等。现任Pacific Journal of Optimization、Statistics、Optimization & Information Computing等期刊编委。主持国度级和省部级项目多项。在Math Program、SIAM J Optim、SIAM J Matrix Anal Appl等顶级期刊颁发论文多篇。
演讲内容简介：
Recovery and forecast of network traffic data from incomplete observed data is an important issue in internet engineering and management. In this paper, by fully considering the temporal stability and periodicity features in internet traffic data, a novel optimization model for internet data recovery and forecast is proposed, which is based upon the newly introduced higher order tensor decomposition form called tubal tensor train decomposition. Moreover, by introducing auxiliary variables and penalty techniques, a relaxation of the proposed model is obtained. Then, an easy-to-operate and effective algorithm for solving the relaxation model is proposed. We prove that the sequence generated by the proposed algorithm converges to a stationary point of the established relaxation model. A series of numerical experiments about the recovery of structurally missing traffic data and the traffic data prediction on the widely used real-world datasets demonstrate that our approach have favorable performance than some state-of-the-art tensor/matrix based approaches.
欢迎宽大师生参与！
