Accelerated Materials Design Driven by Machine Learning
Thursday, March 21, 2024
11:00 am - 12:00 pm
An atomistic-level understanding of the physical properties of materials drives the development of new technologies. Data science and artificial intelligence have emerged as novel tools for understanding and designing materials at the atomistic level. In this presentation, Mingjian Wen will discuss the challenges involved in creating such tools, focusing on the construction of neural network interatomic potential models to study the mechanical and thermal properties of 2D materials and how to quantify the uncertainty in such methods. Wen will also discuss the development of structure–property relationships for inorganic materials, particularly focusing on how to address physical constraints and symmetry requirements inherent in tensorial properties through the utilization of geometric deep learning.
About the Speaker
Mingjian Wen is an assistant professor of Chemical and Biomolecular Engineering in the Cullen College of Engineering and a Presidential Frontier Faculty Fellow at the University of Houston. Wen obtained his Ph.D. in the Department of Aerospace Engineering & Mechanics at the University of Minnesota and worked as a postdoctoral scholar on the Materials Project at Lawrence Berkeley National Lab. His research centers on developing data-driven and machine learning approaches for computational understanding and design of energy materials. He is also interested in open cyberinfrastructure for materials modeling.
- Location
- Zoom
- Cost
- Free
- Contact