Wei Lin, Fudan University
601 Pao Yue-Kong Library
In this talk, I will introduce two model-free frameworks of dynamical time series analytics. One framework is to detect the causation interactions among a large group of dynamical variables, which probably recovers a network hidden in a real-world system we are concerned. The second framework is to make a forecast or future prediction based only on short-term and high-dimensional time series, which is usually believed to be a challenging task. Both frameworks use the advantages of Taken’s embedding techniques, which reveals that utilization of dynamical system theory is more likely to exploit useful information from time series not only from the models but also from the real-world systems.
Dr. Wei Lin received the Ph.D. degrees in applied mathematics from Fudan University, Shanghai, China, in January, 2003 with specialization in dynamical systems, bifurcation and chaos theory. Now, he is a Professor in applied mathematics of Fudan University, China and serving as the Vice Dean of the Institute of Science and Technology for Brain-Inspired Intelligence and as the Director of the Centre for Computational Systems Biology, Fudan University, China. His current research interests include bifurcation and chaos theory, stability and oscillations in hybrid systems, stochastic systems and complex networks, data assimilation, causality analysis, and their applications to systems biology and artificial intelligence. He has his recent works published in the journals including PNAS, PRL, IEEE TAC, SIAM J. Con. Opt., and CHAOS. Dr. Lin is currently the Vice Chair of the Shanghai Society of Nonlinear Sciences, a Board Member of the International Physics and Control Society, an AE of the International Journal of Bifurcation and Chaos, and a member of Editorial Advisory Board of CHAOS. He received the Excellent Young Scholar Fund from NSFC in 2013, and becomes a Highly Cited Chinese Researcher in General Engineering according to Elsevier.