Yu awarded NSF grant to study safety in learning-enabled systems

Posted: October 2, 2023
Smiling woman in front of a gray background

The first time proved to be the charm for ISE Assistant Professor Xian Yu’s submission for a National Science Foundation grant. 

Yu and a team that includes, University of Michigan Professor Lei Ying (electrical engineering and computer science) and Arizona State University Professor Yongming Liu (aerospace and mechanical engineering) and Associate Professor Wenlong Zhang (manufacturing systems and networks), were awarded a $1.5 million grant from NSF. They will study “Safe Distributional-Reinforcement Learning-Enabled Systems: Theories, Algorithms and Experiments.” 

Yu says it is part of a $20 million allocation of funding for Safe Learning-Enabled Systems through NSF’s Directorate for Computer and Information Science and Engineering. 

“This is a collaborative project that spans multiple disciplines, including operations research, computer science, manufacturing and mechanical engineering,” she says. “In this project, we aim to understand the fundamental properties of the risk-sensitive distributional reinforcement learning (DRL) framework.  

“Using the distributional information, what safety and risk guarantees can DRL provide, beyond the traditional reinforcement learning? How can we guarantee global convergence under DRL and what is the fundamental iteration/sample complexity? We will validate this new framework and algorithms in both high-fidelity simulations and real-world experiments using unmanned aerial vehicles.” 

The Safe Learning-Enabled Systems program is a partnership between the NSF, Open Philanthropy and Good Ventures and seeks “to foster foundational research that leads to the design and implementation of learning-enabled systems in which safety is ensured with high levels of confidence,” according to the program synopsis. 

Yu said she discovered the NSF program solicitation last spring and found that it would align well with her research interests. 

“I have been working on the intersection of optimization and reinforcement learning and collaborating with some colleagues from computer science for one year,” she says. “We discussed this opportunity and decided to team up to submit a proposal there. This is an exciting opportunity for me to recruit talented students and work with great colleagues.” 

Earlier this year, Yu was the recipient of the 2023 Pritsker Doctoral Dissertation Award from the Institute of Industrial & Systems Engineers and says she is grateful for the recognition she has received this year and hopes to encourage other students to pursue careers in industrial engineering and operations research. 


Story by Nancy Richison

Category: Faculty