Efficient and Safe Integration Between Learning and Control in Physical Systems

Abstract:  Recent years have witnessed remarkable advances in machine learning and reinforcement learning. These advances stimulate a growing interest in integrating learning technologies into physical systems for better performance, e.g., power systems, robotics, transportation systems. However, many challenges remain unsolved for effective integration of learning and physical systems, such as limited computation, safety issues, scalability, etc.

In this talk, I will present my work on addressing these challenges in the context of online decision-making. Specifically, I will first discuss online optimization and control in time-varying environments with future prediction information. I will introduce fast/gradient-based algorithm design by leveraging the structural properties of this problem and investigate the fundamental limits. Secondly, I will present my work on safe online learning for controlling dynamical systems with model uncertainties. I will provide algorithm design with both safety and non-asymptotic optimality/regret guarantees. Lastly, I will briefly mention my work on addressing other challenges and discuss future research directions.

Bio: Yingying Li is currently a postdoc researcher at the Coordinated Science Laboratory (CSL) and the Department of Industrial and Enterprise Systems Engineering (ISE) at the University of Illinois at Urbana-Champaign (UIUC). She received her Ph.D. at Harvard University in 2021 and her B.S. degree at the University of Science and Technology of China (USTC) in 2015, both in Applied Mathematics. She was also a research intern at MIT-IBM Watson AI Lab in the summer of 2020. Her research interests lie in the intersection of control, machine/reinforcement learning, and optimization, with applications in smart grids, smart cities, and robotics. Her work was selected as the Editor’s Choice by Automatica. She was also selected as a Future Digileader by the KTH Royal Institute of Technology in 2019.