Northwestern Engineering’s Chang-Han Rhee has received a Faculty Early Career Development Program (CAREER) award from the National Science Foundation (NSF), the foundation’s most prestigious honor for junior faculty members. 

Rhee, assistant professor of industrial engineering and management sciences at the McCormick School of Engineering, will receive $568,493 over five years from NSF’s Division of  Civil, Mechanical, and Manufacturing Innovation.

The award supports early career development of individuals who exemplify the role of teacher-scholar through outstanding research, excellent education, and the integration of education and research.

Rhee’s research interests include stochastic simulation, applied probability, experimental design, and machine learning. A former postdoctoral fellow at the Georgia Institute of Technology and the Centrum Wiskunde & Informatica who earned his PhD in computational and mathematical engineering from Stanford University in 2013, Rhee joined the Northwestern Engineering faculty in 2018.

With his CAREER award, titled “Catastrophic Rare Events: Theory of Heavy Tails and Applications,” Rhee will develop mathematical tools that provide strategies to understand and mitigate risk associated with the “heavy-tail phenomena,” a mathematical structure that underlies seemingly disparate rare events such as global pandemic, power system failures, and financial crisis. A particularly well-known and simple manifestation of heavy tails is the “80-20 rule”—for example, the richest 20 percent of the population controls 80 percent of the wealth—whose variations are repeatedly discovered in a wide variety of application areas. Under the presence of heavy tails, high-impact rare events are guaranteed to happen eventually.

“The ongoing global pandemic clearly illustrates that once-a-hundred-year rare events matter,” Rhee said. “Besides the pandemic, there are countless other examples that are rare and impactful at the same time: the 2007 financial crisis, the 2011 tsunami in Japan followed by the nuclear meltdown, and the 2012 blackout in India, to name a few. Moreover, many important problems in science and engineering—such as the generalization mystery of deep neural networks in artificial intelligence—can be formulated as a (less dramatic but just as crucial) rare event analysis problem.

“Remarkably, many of these problems exhibit common mathematical structures: heavy tails (extreme variability), large-scale system dynamics, and abrupt failure/transition of the system," Rhee added. "The objective of this project is to develop a comprehensive set of tools for analyzing heavy-tailed rare events and expanding our understanding of how system failures and phase transitions arise in many stochastic systems and modern algorithms.”

Building on the mathematical machinery developed in this NSF project, Rhee will address open problems in artificial intelligence and insurance risk management. In particular, Rhee aims to provide rigorous theoretical foundation for designing reliable and accountable AI so that the technology can be applied to high-stake decision-making problems with confidence. Rhee will also develop an educational program to broaden STEM interest in underrepresented communities and train future academic, industry, and government leaders by equipping them with fundamental skills in risk analysis.