TYC Highlight Seminar: Machine-Learned Force Fields for Molecular Simulations Beyond AlphaFold and Empirical Potentials

TYC Highlight Seminar: Machine-Learned Force Fields for Molecular Simulations Beyond AlphaFold and Empirical Potentials Share on X

Alexander Tkatchenko, University of Luxembourg

The convergence between accurate quantum-mechanical (QM) models (and codes) with efficient machine learning (ML) methods seem to promise a paradigm shift in all-atom simulations. Many challenging applications are now being tackled by increasingly powerful QM/ML methodologies (https://doi.org/10.1021/acs.chemrev.0c01111; https://doi.org/10.1021/acs.chemrev.1c00107). These include modeling covalent materials, molecules, molecular crystals, surfaces, and even whole proteins under physiological conditions (https://www.science.org/doi/abs/10.1126/sciadv.adn4397; https://doi.org/10.1021/jacs.5c09558).

In this talk, I will attempt to provide a reality check on these recent advances and on the developments required to enable fully predictive dynamics of complex functional (bio)molecular and material systems. Multiple challenges are highlighted, in particular transferability in chemical space and interatomic interactions that should enable this field to grow for the foreseeable future.

Add to calendar:

Google Calendar

iCalendar