Advances in Probabilistic Generative Modeling for Scientific Machine Learning

Abstract

Leveraging large-scale data and computing accelerator systems, statistical learning has led to significant paradigm shifts in many scientific disciplines. Grand challenges in science have been tackled with exciting synergy between disciplinary science, physics-based simulations via high-performance computing, and powerful learning methods.

In this talk, I will describe several vignettes of our research on modeling complex dynamical systems characterized by partial differential equations with turbulent solutions. I will also demonstrate how machine-learning technologies, especially advances in generative AI, are effectively applied to address the computational and modeling challenges in such systems, exemplified by their successful applications to weather forecasting and climate projection. I will also discuss the new challenges and opportunities that future machine-learning research faces.

Bio

Fei Sha is an AI Research Scientist at Meta. He is broadly interested in probabilistic modeling, uncertainty quantification, dynamical systems, and probabilistic reasoning in LLMs. Before joining Meta, he led a team of scientists and engineers at Google Research, working in various topics, incuding basic methods and technology for LLMs, probabilistic generative modeling and their applications to dynamical systems (such as weather and climate). Before joining Google Research, he was a Professor of Computer Science and the Zohrab A. Kaprielian Fellow in Engineering at the University of Southern California (USC). He has been recognized with numerous awards and accolades for his innovative work, including being selected as an Alfred P. Sloan Research Fellow in 2013 and receiving an Army Research Office Young Investigator Award in 2012. He has a PhD in Computer and Information Science from the University of Pennsylvania and BSc and MSc degrees from Southeast University (Nanjing, China).