[Report]Probabilistic Reduced-Dimensional Vector Autoregressive Modeling of Data with Low Dynamic Dimensions

April 27,10:00-11:00

Editor:College of Information Science and Technology Time:2024-04-23

SpeakerS. Joe Qin

TimeApril 27, Saturday, 10:00-11:00am

VenueConference Center 3rd floor lecture hall, East campus

Abstract:

In this talk I will present a novel latent vector autoregressive framework to model reduced dimensional dynamic components. High dimensional time series data are common in modern engineering, internet of things, financial, economic, autonomous, and control systems. Dimension reduction and dynamics modeling tasks are simultaneously required, but traditional multivariate time series analysis does not handle them simultaneously. We present a probabilistic reduced-dimensional vector autoregressive (PredVAR) model to extract low-dimensional dynamics with a canonical correlation analysis (CCA) objective. The dynamic latent variable scores are enforced with a reduced dimensional VAR model with maximized predictability. The model utilizes an oblique projection to partition the measurement space into a subspace that contains the reduced-dimensional dynamics and a complementary static subspace. We develop an iterative PredVAR algorithm using maximum likelihood and the expectation-maximization (EM) framework. Time-dependent data from a chaotic Lorenz oscillator and an industrial process are used to test the superiority of the proposed algorithm. The reduced-dimensional latent dynamic modeling framework has potentially wide applications in prediction, control, and diagnosis of anomalies.

Biography:

Professor Qin Sizhao was admitted to Tsinghua University in 1979 and later received a full scholarship to pursue a doctoral degree at the University of Maryland in the United States. He has served as Vice Dean of the School of Engineering at the University of Southern California, Vice President of the Chinese University of Hong Kong (Shenzhen), Founding Dean of the School of Data Science at City University of Hong Kong, and current President of Lingnan University in Hong Kong.

Professor Qin is a member of the National Academy of Inventors in the United States, the European Academy of Sciences and Arts, the Hong Kong Academy of Engineering Sciences (HKAES), the International Federation of Automatic Control, the American Society of Chemical Engineers (AIChE), and the Institute of Electrical and Electronics Engineers (IEEE). In 2022, he was awarded the AIChE Computational and Systems Technology (CAST) Computational and Chemical Engineering Award, and the IEEE Control Systems Society Technology Conversion Award in the same year. He is the first and only scholar in Greater China to receive both awards simultaneously. In addition, Professor Qin has won multiple awards in the development of his research career, including the National Science Foundation CAREER Award in the United States, the Northrop Grumman Best Teaching Award from the Viterbi School of Engineering in 2011, the DuPont Young Professor Award, the Haliburton/Brown&Root Outstanding Young Scholar Award from the National Natural Science Foundation of China, the Changjiang Lecture Professor from the Ministry of Education, and the IFAC Best Paper Award for a Model Predictive Control paper published in Control Engineering Practice.

Professor Qin has 12 American patented inventions and has served as a senior editor for the Journal of Process Control, editor for Control Engineering Practice, and editorial board member for the Journal of Chemometrics. He was the Chairman of the Organizing Committee of the 10th IFAC Advanced Control Seminar on Chemical Processes (ADCHEM 2018). His research interests include data science and analysis, machine learning, process monitoring, model predictive control, system recognition, intelligent manufacturing, smart cities, and health predictive maintenance.