报告人:Professor S. Joe Qin
Lingnan University, Hong Kong SAR
报告时间、地点:15:00-16:30, 10月11号, 2024;9001w以诚为本入口B222
摘要:
Multi-dimensional time series are ubiquitous in engineering, science, and economics. While the dimension of sensors increases with modern sensing technology and data acquisition, the dimension of dynamics is often relatively small. In this talk I will present a probabilistic latent vector autoregressive framework, where dimension reduction and optimal dynamic prediction are simultaneously achieved. The dynamic latent variables are enforced by a reduced dimensional predictive model with maximized predictability. The solution requires an oblique projection to achieve uncorrelated realizations of noises in the dynamic and static subspaces. Counter intuitive insight is revealed in this reduced-dimensional formulation, which can shed light on deep learning extensions. An iterative solution is developed using a maximum likelihood framework. Data from a chaotic Lorenz oscillator and industrial processes are used to show the superiority of the proposed algorithms. The reduced-dimensional dynamic modeling framework has potentially wide applications in prediction, control, and diagnosis of anomalies.
报告人简介:
Joe Qin is the Wai Kee Kau Chair Professor of Data and President of Lingnan University in Hong Kong. He obtained his B.S. and M.S. degrees in Automatic Control from Tsinghua University in Beijing and his Ph.D. degree in Chemical Engineering from University of Maryland at College Park. Qin’s research interests include data science and analytics, statistical and machine learning, industrial AI, process monitoring, model predictive control, system identification, smart manufacturing, smart cities, and smart energy management. Dr. Qin is a member of the European Academy of Sciences and Arts and Fellow of the Hong Kong Academy of Engineering Sciences, the U.S. National Academy of Inventors, IFAC, AIChE, and IEEE. He is the recipient of the 2022 AIChE CAST Computing Award, 2022 IEEE CSS Transition to Practice Award, the U.S. NSF CAREER Award. His h-indices for Web of Science, SCOPUS, and Google Scholar are 69, 76, and 88, respectively.