Sam D
- Research Program Mentor
PhD at Harvard University
Expertise
Soft Matter Physics, Learning in Physical Systems, Machine Learning, Friction, Granular Media
Bio
Sam is a postdoctoral researcher at the University of Pennsylvania, studying the intersection between physics and machine learning (ML), specifically how the two fields can inform and support each other. He's currently building physical systems that can learn without the help of a computer. He also studies behaviors of complex systems with less buzz, like sand. Outside of work, Sam gets on stage and makes stuff up to make people laugh. He's been doing improvisational theater, including teaching, directing, and performing, for over a decade. He's also a reluctant but consistent runner, perpetually wants to play basketball, and is always up for a new sport - lately he's tried badminton, squash, and disc golf.Project ideas
Hijacking Physics to do Math for Us
We use math to do a lot of things, like run computers or make predictions. We also use math to describe physical behaviors in the world. But that also means that physical behaviors in the world "do math." In a sense, the world around us is constantly doing "calculations" with physics. In this project, we'll figure out a way to get the world to do our math for us, either in simulation or a simple physical system. We'll pick an example task (measure vibration/seismic activity over time, sense changes in shape, detect humidity, etc), and figure out how to make a reliable test without a computer. We'll think about experimental design, dealing with the noisiness of the real world, and critical data analysis.