Berlin 2018 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

SYMS: Symposium Data-driven Methods in Molecular Simulations of Soft-Matter Systems

SYMS 1: Data-driven Methods in Molecular Simulations of Soft-Matter Systems

SYMS 1.2: Hauptvortrag

Montag, 12. März 2018, 15:30–16:00, H 0105

A Generally-Applicable Machine-Learning Scheme for Materials and Molecules — •Michele Ceriotti — Institute of Materials, EPFL, Lausanne, Switzerland

Determining the stability of molecules and condensed phases is the cornerstone of atomistic modelling, underpinning our understanding of chemical and materials properties and transformations. I will show that a machine-learning model, based on a local description of chemical environments and Bayesian statistical learning, provides a unified framework to predict atomic-scale properties. It captures the quantum mechanical effects governing the complex surface reconstructions of silicon, predicts the stability of different classes of molecules with chemical accuracy, and distinguishes active and inactive protein ligands with more than 99% reliability. The universality and the systematic nature of this framework provides new insight into the potential energy surface of materials and molecules. I will also discuss how the method can be extended to yield a "symmetry-adapted" Gaussian process regression approach that is capable of learning tensorial properties without the need of defining explicitly a local reference frame.