The Eric and Wendy Schmidt AI in Science Postdoctoral Fellowship

Peter Lu

Eric and Wendy Schmidt AI in Science Postdoctoral Fellow



In the physics community, there has been a growing interest in using modern machine learning (ML) approaches to tackle difficult problems in experimental data analysis as well as address important open questions in theoretical physics. However, despite the significant advances in ML over the past decade, researchers are still in the process of learning how to use these methods in scientific applications. To solve hard problems in physics and other scientific fields, researchers must improve the interpretability, generalization performance, and data efficiency of ML methods by creating new approaches that are specifically adapted for scientific applications and designed to provide insight into the underlying physical systems. Peter Lu’s research focuses on addressing these issues by (1) creating ML architectures and algorithms that incorporate physical laws and constraints, (2) integrating these physics-informed architectures with existing computational methods, and (3) developing interpretable representation learning methods for discovering meaningful physical features that help us understand and characterize unknown physical systems.


Peter Y. Lu is an Eric and Wendy Schmidt AI in Science Postdoctoral Fellow at the University of Chicago working at the intersection of physics and machine learning. He received a Ph.D. in Physics from MIT in 2022, where he was an NDSEG Fellow, and an A.B. in Physics and Mathematics from Harvard in 2016. His research interests include physics-informed machine learning and interpretable representational learning with applications in nonlinear dynamics, condensed matter physics, photonics, fluid dynamics, biophysics, and other areas. Peter Lu aims to develop new computational methods for modeling and understanding physical systems with an emphasis on incorporating physics-informed priors and identifying relevant and interpretable latent representations.


Scroll to Top