Chemistry departmental seminar — Dr. Sergei Tretiak
About the event
Title: Machine learning for materials and chemical dynamics
Abstract: Machine learning (ML) is quickly becoming a premier tool for modeling chemical processes and materials. Generally, ML provides a surrogate model trained on the dataset of some reference data. This model establishes a relationship between structure and underlying chemical properties, guiding chemical discovery. Designing high-quality training data sets is crucial to overall model accuracy. To address this this problem, I will describe the active learning strategy, in which new data are automatically collected for atomic configurations that produce large ML uncertainties. The locality approximation underpinning favorable computational scaling of the ML models, is another severe limitation that fails to capture long-range effects that may arise from charge transfer, polarization, electrostatic or dispersion interactions. I will also discuss how ML models can overcome nonlocality (via introduction of interaction layers, self-consistent cycles, or charge equilibration schemes) and exemplify their performance for chemical problems with nonlocalities. All these advances are exemplified by applications to molecules and materials. Exciting new method development and explosive growth of user-friendly ML frameworks, designed for chemistry, demonstrate that the field is evolving towards physics-based models augmented by data science.
Bio: Sergei Tretiak is a T-1 deputy group Leader in the Theoretical Division at Los Alamos National Laboratory (LANL) and a Los Alamos National Laboratory Fellow. He received his Master’s degree in Physics in 1994 from Moscow Institute of Physics and Technology (Russia) and his Chemistry doctorate in 1998 from the University of Rochester (US). He was then a Director-funded postdoctoral fellow (1999-2001), and subsequently became a staff scientist at LANL and a member of the DOE-funded Center for Integrated Nanotechnologies (CINT). Tretiak also serves as Adjunct Professor at the University of California, Santa Barbara (UCSB) (2015-present) and at Skolkovo Institute of Science and Technology, Skoltech, Russia (2013-present). He became an American Physical Society Fellow (APS) in 2014 and a Fellow of the Royal Society of Chemistry, (RSC) in 2019. He has also received the Humboldt Research Award (2021), the Los Alamos Postdoctoral Distinguished Mentor Award (2015) and the Los Alamos Fellow’s Prize for Research (2010). His research interests include development of electronic structure methods for molecular optical properties, nonlinear optical response of organic chromophores, non-adiabatic dynamics of electronically excited states, optical response of confined excitons in conjugated polymers, carbon nanotubes, semiconductor nanoparticles, mixed halide perovskites and molecular aggregates, the use of Machine Learning and Data Science toward modeling electronic and chemical properties. Tretiak has published more than 300 scientific publications cited more than 20,000 times (h-index=70, WebOfSci) and he has presented more than 300 invited and keynote talks in the US and abroad.