UVA Data Points Podcast: Extreme Physics

Animation of extreme physics
Visualization of an extreme physics event

In this episode of UVA Data Points, we explore how data science is helping researchers simulate and understand some of the most extreme physical events on Earth, from floods in Texas to hypersonic flight. Our guests are Stephen Baek, a leading expert in geometric deep learning and an associate professor of data science at the University of Virginia, and Jack Beerman, a Ph.D. student whose work is shaping real-world applications.

Together they discuss how artificial intelligence is transforming fields like weather forecasting, materials design, sports performance and military innovation, and why graduate researchers like Beerman are essential to moving this work forward.


Extreme Physics

At the Visual Intelligence Laboratory, Stephen Baek leads a team interested in developing computer vision algorithms to predict those interesting features, not only learned from data but also informed by physics equations.

This animation shows many real-world physics phenomena creating interesting visual patterns. Especially for extreme physics phenomena, these visual patterns may include many interesting features such as boundaries with sharp contrasts, small protruding features, spikes/gushes/surges that rapidly appear and then suddenly disappear, etc. 
How do tennis players generate a massive force during serve? This animation displays some of the over 6,000 tennis serve motions during the 2024 U.S. Open tennis tournament, automatically collected and analyzed using AI (see this dataset for more). The Visual Intelligence Laboratory uses computer vision to capture the biomechanics of elite athletes in different sports and models their extreme dynamics using physics-aware machine learning algorithms.

Author

Creative Manager