October 16, 2020, Zoom

In this talk, we discuss some recent work on the connections between machine learning and dynamical systems. These come broadly in three categories, namely machine learning via, for and of dynamical systems, and here we will focus on the first two. In the direction of machine learning via dynamical systems, we introduce a dynamical approach to deep learning theory with particular emphasis on its connections with optimal control. In the reverse direction of machine learning for dynamical systems, we discuss the approximation and optimization theory of learning input-output temporal relationships using recurrent neural networks, with the goal of highlighting key new phenomena that arise in learning in dynamic settings.

Qianxiao Li is an assistant professor in the Department of Mathematics, National University of Singapore and a research scientist in the Institute of High Performance Computing, A*STAR. He graduated with a BA in mathematics from University of Cambridge in 2010, and a PhD in applied mathematics from Princeton University in 2016, after which he joined the institute of high performance computing as a research scientist. In 2019, he joined the department of mathematics at NUS as an assistant professor, while holding a joint appointment with A*STAR. His research interests include theoretical machine learning, numerical analysis, optimal control, and the application of data-driven methods to problems in the natural sciences.