Guest Seminar - Accelerating Materials Design via Computation and Machine Learning Combined Approaches

3:30pm - 4:30pm
Room 1409 (Lift 25-26)

A long-standing challenge for materials design is to clarify materials microscopic features, with providing their linkage to macroscopic properties and processing parameters. With the recent development of computing equipment and algorithms, modeling and simulations, assisted by artificial intelligence techniques, have become a promising means to tackle the above problem. In this presentation, computation-based strategies for accelerating materials design will be discussed, through the demonstration of three examples from different aspects on combining modeling and machine learning. In the first example, an effective ensemble learning framework was proposed, for automated identification on atomic crystal structures from molecular dynamics trajectories, as a generalizable tool to extract materials descriptors. In the second example, Bayesian optimization guided coarse-grained molecular dynamics simulations were performed, to systematically investigate the effects of molecular level material properties on lithium-ion transport in polymeric material systems, for promoting an efficient screening of solid polymer electrolytes. In the last example, a three-dimensional generative-adversarial-networks-based model was established, as an attempt toward inverse design of composite materials with complex structures and multiple optimization objectives.

Event Format
Speakers / Performers:
Yanming Wang

Yanming Wang is currently an associate professor at the University of Michigan-Shanghai Jiao Tong University Joint Institute. He also serves as the executive director of the Center for Perovskite Photovoltaics Research, and an executive board member of the Future Battery Research Center at the Global Institute of Future Technology of Shanghai Jiao Tong University (SJTU). He received B.E. in Materials Science and Engineering from Shanghai Jiao Tong University. After that, he attended Stanford University for his M.S. and Ph.D. in Materials Science and Engineering. Before joining SJTU, he was a postdoctoral researcher at Massachusetts Institute of Technology. His research aims to combine material modeling and machine learning, to guide the design and realization of both novel organic and inorganic materials, for energy and electronic applications.

Language
English
Recommended For
Faculty and staff
PG students
Organizer
Department of Chemical & Biological Engineering
Post an event
Campus organizations are invited to add their events to the calendar.