Abstract
Control and machine learning are two high-impact areas. On one hand, control has been a crucial part of modern engineered systems such as commercial aircraft, space vehicles, and nuclear plants. On the other hand, machine learning has recently revolutionized the field of artificial intelligence (AI) and led to state-of-the-art results in computer vision, natural language processing, and Go. The developments of next generation intelligent systems such as self-driving vehicles, humanoid robotics, smart buildings, and automated healthcare require a rapprochement of these two areas. In this talk, we will discuss the interplay between control theory and machine learning, showing that the techniques used by each side can be explored to impact the other side. The first half of this talk focuses on “control for learning.” Specifically, we will present a control-theoretic approach for designing large-scale Lipschitz neural networks achieving state-of-the-art certified robustness on image classification tasks. The second half of this talk focuses on “learning for control.” Specifically, we will discuss the theoretical guarantees of policy-based reinforcement learning (RL) techniques on linear robust control problems including H-infinity state-feedback synthesis and mixed H2/H-infinity control design. Finally, we will provide some remarks on open problems at the intersection of learning and control.
Biography
Bin Hu received the B.Sc. in Theoretical and Applied Mechanics from the University of Science and Technology of China in 2008, and received the M.S. in Computational Mechanics from Carnegie Mellon University in 2010. He received the Ph.D. in Aerospace Engineering and Mechanics at the University of Minnesota in 2016. Between July 2016 and July 2018, he was a postdoctoral researcher in the Wisconsin Institute for Discovery at the University of Wisconsin-Madison. He is currently an assistant professor in the Department of Electrical and Computer Engineering at the University of Illinois Urbana-Champaign and affiliated with the Coordinated Science Laboratory. His research focuses on building fundamental connections between control theory and machine learning. He received the NSF CAREER award and the Amazon research award in 2021.
