91����

91���� is hosting 10 undergraduate students from across the country this summer on Mies Campus to participate in data and computational mathematics research projects to explore how data is used to solve the world’s biggest problems.

The students, who attend nine different universities from across the country, have converged at 91���� to participate in the Summer Undergraduate Research Experience, a 10-week, intensive research program funded by the National Science Foundation’s Research Experience for Undergraduates program. 

The students work side by side with 91���� faculty and doctoral students to dive into real-world research that addresses today’s most pressing data science and computational mathematics challenges.

“The biggest transformation over the last 15 or 20 years has been from solving problems based on logic to solving problems based on data,” says Fred Hickernell, professor of applied mathematics at 91����. “We use that data to understand the world around us, from large-language models to physical objects.”

Hickernell will lead one team of students, along with Senior Research Associate Nathan Kirk, through a project titled “Speedier Simulations,” which will immerse students in quasi-Monte Carlo methods and the low-discrepancy points that make it work. Using machine-learning tools such as graph neural networks, the team will add state-of-the-art procedures to distribute data points more evenly and achieve more accurate results. The team also will explore new applications for quasi-Monte Carlo methods, new randomization techniques for machine learning-generated low-discrepancy points, and the performance of these new low-discrepancy sets.

Another team will explore how artificial intelligence, machine learning, and the Internet of Things can work together to prevent heat stroke in a project titled “Enhancing Construction Worker Safety Through AI,” which is being led by Sou-Cheng Choi. The team will examine the specific physiological and mental states of individual workers, as well as the dynamic environmental factors that contribute to hazards. Wearable AI technology presents new opportunities for predicting and mitigating workplace incidents and to explore the application of AI in enhancing safety within the construction industry. AI models are used to predict risky worker mental states by processing and analyzing time series of physiological data and external factors, including weather conditions and site-specific hazards, with the data being collected through construction helmets.

Huiling Liao, assistant professor of applied mathematics, will lead another team of students to work with real-world datasets—such as environmental measurements, traffic accident reports, or public health records—and apply statistical and computational methods to explore how changes unfold over time and across regions in a spatial-temporal modeling project. It involves analyzing data, such as air pollution levels, disease incidence, or weather patterns, that vary across both space and time to uncover patterns, make predictions, and understand dynamic processes, patterns, and uncertainty quantification.

College of Computing Dean Lance Fortnow notes how each project has at least an element of AI incorporated into the research.

“You’ll find ways to use AI—not to do the project, because that would be a mistake—but as a companion,” he said to the researchers at a kickoff event. “I view AI as a tool to help me. I hope you find ways to make it a powerful tool for you.”

Photo: SURE students and faculty gathered in classroom.