Dresden 2020 – wissenschaftliches Programm

Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...

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

KFM: Fachverband Kristalline Festkörper und deren Mikrostruktur

KFM 15: Postersession KFM

KFM 15.29: Poster

Donnerstag, 19. März 2020, 16:00–18:30, P2/1OG

Machine Learning in VASP — •Ferenc Karsai¹, Ryosuke Jinnouchi², and Georg Kresse² — ¹VASP Software GmbH, Sensengasse 8, Vienna, Austria — ²University of Vienna, Department of Physics, Sensengasse 8, Vienna, Austria

An efficient and robust on-the-fly machine learning force field method implemented into the Vienna Ab-initio Simulation Package (VASP) is presented. This method realizes automatic generation of machine learning force fields on the basis of Bayesian inference during molecular dynamics simulations, where the first-principles calculations are only executed, when new configurations out of already sampled data sets appear. The power of the method is demonstrated on several applications such as e.g. melting points of ionic and covalent compounds and solid-solid phase transitions in perovskites. The applications show that 99% of the ab-initio calculations are skipped. This way the calculations are accelerated by more than 3 orders of magnitude, while still being able to quantitatively reproduce the ab-initio results. The implementation of our on-the-fly learning scheme is fully automatized and is mainly controlled by a few parameters. This way the amount of human intervention for the usually laborious task of training is hugely reduced.