Meet The Postdoc is a monthly series highlighting the AI Lab’s postdoctoral researchers.

Ni “Jenny” Zhan is a postdoctoral researcher for AI for Accelerating Invention, advised by Ryan Adams. She works on artificial intelligence for design and simulation of dynamic processes and materials. She holds a doctorate in chemical engineering from Carnegie Mellon University.

What brought you to the AI Lab?

I wanted to do more in depth AI/ML research following my Ph.D. with Ryan Adams, who is co-director of AI for Accelerating Invention. Particular research topics include AI for accelerating materials simulation, design of scientific experiments, solving inverse design problems and more.

What was your Ph.D. thesis about? 

My Ph.D. thesis was titled, “Machine learning models and uncertainty for atomic simulations.” We use molecular dynamics simulation of atomic systems to learn about material properties and behavior. Machine learned models are used in the simulations as high accuracy surrogates at much faster computational cost. I discovered a novel relationship between diffusion and viscosity in liquid aluminum silicon and described an uncertainty quantification method that determines when machine learned models extrapolate.

What has the experience at the AI Lab been like?

I very much enjoy my experience at the AI Lab, and am grateful for the funding support. Through events and my research work at the AI Lab, I’ve been able to meet and join collaborations with researchers interested in AI across Princeton and other institutions. We’ve worked on interesting topics including symmetrization of neural network wave function solvers, improving the coding ability of language models, generative models for crystals, etc.

How has your postdoc experience differed from your Ph.D.?

I had great collaborations during my Ph.D. and did in-depth molecular dynamic studies of liquid alloys and ML uncertainty quantification. During my postdoc, I have gotten to do more collaborations, especially with researchers of different backgrounds including computer science, physics and materials science. My postdoc has expanded my experiences with developing AI models for physics and new material systems.

Tell us about your research you’re working on. How does it contribute to solving real-world problems?

Currently I’m working on integrating machine learning with methods to solve Schrödinger’s equation at high accuracy. I developed a way to simulate systems exhibiting exotic magnetic phases and spin-orbit coupling. We learn unknown physics that could be developed into quantum devices such as increased energy efficient transistors. My research more broadly helps us discover and design new materials such as catalysts, alloys, and fuel cells.

How do you think the experience at the AI Lab will advance your career goals?

My time at the AI Lab expanded the scope and depth of my research experience and refined my career priorities.

Do you have any advice for someone just starting a postdoc?

Refine your research interests early and plan projects to target them.

What do you like to do outside of work?

I like practicing yoga and piano, visiting new places, and having a variety of life experiences. 

What’s your favorite snack at the AI Lab?

Potato chips.