More Than Physics, More Than Data: Integrated Machine-Learning Models for Chemistry

Machine-learning techniques are often applied to perform “end-to-end” predictions, making a black-box estimate of a property of interest using only a coarse description of the corresponding inputs. In contrast, atomic-scale modeling of matter is most useful when it allows a mechanistic insight into the microscopic processes that underlie the behavior of molecules and materials. Dr. Ceriotti will provide an overview of the progress that has been made combining these two philosophies. He will discuss several examples of the application of these ideas, from the calculation of excited states of molecules to the design of high-entropy alloys for catalysis, emphasizing both the accuracy and the interpretability that can be achieved with a hybrid modeling approach.

Dr. Ceriotti leads the laboratory for Computational Science and Modeling in the Institute of Materials at EPFL. He is one of the core developers of several open-source software packages, including iPi and Chemiscope, and he serves the atomistic modeling community as an associate editor of the Journal of Chemical Physics, as a moderator of the physics.chem-ph section of the arXiv, and as an editorial board member of Physical Review Materials.

This seminar is part of the Hirschfelder Visitor Program. It is sponsored by the Theoretical Chemistry Institute, Data Science Institute, and the Departments of Materials Science & Engineering and Chemical & Biological Engineering.