Title: Learning and Adaptation in Uncertain Dynamical Systems

Time: Thursday, October 22, 2026, 15:00–17:00

Venue: Second Lecture Hall, New Main Building, Xueyuan Road Campus, Beihang University

Speaker: Guo Lei

Abstract:

Learning and adaptation are fundamental to the modeling and control of uncertain dynamical systems, and serve as a cornerstone of intelligent systems. Statistical learning underpins many advances in machine learning, yet its theoretical foundations typically built on idealized assumptions about data — most notably, that samples are independent and identically distributed (i.i.d.). Such assumptions generally do not hold in dynamical systems, particularly under feedback control, where input–output data are often strongly time-correlated and exhibit inherent non-stationarity.

Therefore, there is an urgent need for more general theoretical frameworks capable of capturing the complex characteristics intrinsic to dynamical system data. This talk will address these problems and challenges, presenting recent research advances from the speaker's team on learning and adaptation inuncertaindynamical systems.

Abstract:

Guo Lei is a research professor at the Academy of Mathematics and Systems Science, Chinese Academy of Sciences (CAS). He served as Director of the Institute of Systems Science at CAS from 1999 to 2002 and as President of the Academy of Mathematics and Systems Science from 2003 to 2012, and currently directs the CAS National Center for Mathematics and Interdisciplinary Sciences.

He was elected an IEEE Fellow in 1998, a Member of the Chinese Academy of Sciences in 2001, and a Fellow of the Academy of Sciences for the Developing World in 2002. In 2007 he was elected a Foreign Member of the Royal Swedish Academy of Engineering Sciences and an IFAC Fellow, and in 2014 he received an honorary doctorate from KTH Royal Institute of Technology. He has delivered plenary lectures at three IFAC World Congresses (1999, 2014, and 2026) and an invited lecture at the International Congress of Mathematicians (2002), and was named a Distinguished Lecturer of the IEEE Control Systems Society in 2012.

His research interests include stochastic systems, system identification, machine learning, adaptive control and filtering, adaptive game theory, control of uncertain nonlinear systems, multi-agent systems, and game-based control systems. He received the IFAC Young Author Award in 1993, the IFAC Outstanding Service Award in 2011, and the IEEE Control Systems Society Hendrik W. Bode Lecture Prize in 2019.

Editor: Lyu Xingyun