Berlin 2024 – wissenschaftliches Programm

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

MM: Fachverband Metall- und Materialphysik

MM 29: Poster II

MM 29.2: Poster

Dienstag, 19. März 2024, 17:00–19:00, Poster B

RuNNer 2.0: An Efficient and Modular Program for Training and Evaluating High-Dimensional Neural Network Potentials — •Alexander L. M. Knoll^{1, 2}, Moritz R. Schäfer^{1, 2}, K. Nikolas Lausch^{1, 2}, Moritz Gubler³, Jonas A. Finkler³, Emir Kocer^{1, 2}, Alea Miako Tokita^{1, 2}, Tsz Wai Ko⁴, Marco Eckhoff⁵, Gunnar Schmitz^{1, 2}, and Jörg Behler^{1, 2} — ¹Theoretische Chemie II, Ruhr-Universität Bochum, Germany — ²Research Center Chemical Sciences and Sustainability, Research Alliance Ruhr, Germany — ³Department of Physics, Universität Basel, Basel, Switzerland — ⁴Department of NanoEngineering, University of California, San Diego, CA, USA — ⁵ETH Zürich, Laboratorium für Physikalische Chemie, Zürich, Switzerland

Machine learning potentials (MLPs) have become a popular tool for large-scale atomistic simulations in chemistry and materials science. They provide efficient access to highly accurate potential energy surfaces (PES) generated from ab initio reference calculations. As methods in this field are becoming more and more complex and reach maturity, the development of efficient and user-friendly tools is increasingly important. We present the second major release version of RuNNer, an open source, stand-alone software package for the construction and evaluation of second-, third-, and fourth-generation high-dimensional neural network potentials (HDNNPs). RuNNer 2.0 unifies the entire workflow in a fully MPI-parallel program: from the generation of atomistic descriptors, over the training of a specific machine learning model, to its final application in molecular dynamics.

Keywords: Software; Implementation; Machine Learning Potentials; Fortran; High-Dimensional Neural Network Potentials