Speaker: Xiao Luo, Ph.D. IDRE Fellow Department of Computer Science University of California Los Angeles      Time: 11:30 AM – 12:30 PM (PST) Date: March 28, 2024 Venue: RSVP here for the Zoom link

Abstract: Many real-world systems, such as disease transmission, molecular dynamics, and spring systems, can be considered as multi-agent dynamical systems, where multiple objects interact with each other and exhibit complex behavior over time. However, modeling and analyzing these systems for scientific applications face several challenges. Firstly, capturing the intricate spatio-temporal correlations in these systems is difficult. Secondly, these models often lack robustness to distribution shifts in real-world scenarios. Thirdly, a gap exists between machine learning models and the specific needs of scientific applications. This talk will describe my research on interacting dynamics system modeling for scientific problems. I will start by discussing my work on graph ODEs for efficiently capturing continuous high-order correlations. Then, I will elaborate on how to address different distribution shifts to improve the generalization. The talk will also explore the practical applications of our models in various fields, including health, molecular biology, genomics, physics, and meteorology. Finally, I will outline future research directions in the field of dynamical system modeling.

About the speaker: Dr. Xiao Luo is a postdoctoral researcher at UCLA’s Department of Computer Science. Previously, he received a B.S. degree in Mathematics from Nanjing University, Nanjing, China, in 2017 and a Ph.D. in the School of Mathematical Sciences from Peking University, Beijing, China in 2022. His research interests include machine learning on graphs, dynamical systems, statistical models, and AI for Science.