Decoding the Glass Genome by Modeling and Machine Learning

Mathieu Bauchy, Assistant Professor
Civil & Environmental Engineering Department, UCLA

From telescope lenses to light bulbs, glass has played a pivotal role in human history. Today, optical glass fibers and touch-screen display glasses have changed the way humans interact with information. Glass also shows great promises to solve some of tomorrow’s grand challenges in healthcare or environment. However, addressing these challenges requires the discovery of new glass formulations exhibiting unusual properties. Rather than relying on traditional trial-and-error Edisonian approaches, accelerating the discovery of new glasses requires one to decode the “glass genome,” that is, to decipher how the properties of glasses are controlled by their underlying composition and structural features (i.e., the glass “genes”). In this presentation, I will discuss the importance of glass in modern society and present some of our recent effort in decoding the glass genome by means of atomic simulations, topological modeling, and machine learning.

About the Speaker

Mathieu Bauchy is an Assistant Professor in the Civil & Environmental Engineering Department at the University of California, Los Angeles. He received his undergraduate education in physics at Ecole Normale Supérieure (Paris, France) before pursuing a Ph.D. in condensed matter at Université Pierre et Marie Curie (Paris). He then joined the Massachusetts Institute of Technology as a postdoctoral associate. Professor Bauchy’s research focuses on revealing the physics governing the behavior of disordered materials by means of multi-scale simulations and machine learning. He has received the Norbert J. Kreidl Award by the American Ceramics Society, which recognizes research excellence in glass science. He has delivered more than 60 scientific presentations and published more than 100 papers.