Engineering News

Zhaoran Wang Receives Prestigious NSF CAREER Award

He will use the award to advance the field of principled deep reinforcement learning

Northwestern Engineering’s Zhaoran Wang has received the Faculty Early Career Development (CAREER) Award from the National Science Foundation (NSF), the foundation’s most prestigious honor for junior faculty members.

The CAREER Award is designed to support promising young faculty members who exemplify the role of teacher-scholar through the combination of outstanding research and education.

Wang, assistant professor of industrial engineering and management sciences and (by courtesy) computer science in the McCormick School of Engineering, will receive $500,000 over five years from NSF’s Division of Electrical, Communications, and Cyber Systems for his project titled “Principled Deep Reinforcement Learning for Societal Systems.”

Zhaoran WangWang, whose research interests include machine learning, optimization, statistics, game theory, and information theory, will explore deep reinforcement learning (RL), a subfield of machine learning that presents new opportunities to control complex and unknown systems using large data inputs. While deep RL algorithms are used extensively to power video games, the systems are inefficient and untrustworthy at scale, making it difficult to incorporate into critical societal domains like healthcare, transportation, and supply chains.

To address these challenges, Wang will establish a theoretical framework for analyzing the computational efficiency and sample efficiency of single-agent deep RL and an algorithmic framework for achieving such efficiencies. Moreover, he will develop a stochastic game framework for achieving safety, robustness, scalability, fairness, risk-awareness, and incentivization in social systems using multi-agent deep RL.

“This CAREER Award would not have been possible without the support and mentorship from my colleagues in the industrial engineering department, especially Chair David Morton, and my mentor, Professor Jorge Nocedal,” Wang said. “This award will build a unified framework that integrates data science and decision science, which have been separately studied in operations research, computer science, and statistical science. I am excited about this new area of data-driven decision science that will emerge from such integration.”

A member of the Center for Deep Learning and the Center for Optimization and Statistical Learning, Wang will also use the award to create education programming that teaches data-driven decision making as a fundamental skill for future generations. Specifically, the initiatives will promote data-driven social leadership and support underrepresented minority researchers and students who face challenges in societal systems, from K-12 education to graduate training. The outreach plan also involves a series of programs held remotely due to the COVID-19 pandemic, including online seminars on data science and artificial intelligence, intern mentorship, and remote engagement with students via programs like DataFest and Client Project Challenge.