RuNNer 2.0: A software suite for high-dimensional neural network potentials
The Journal of Chemical Physics · 21.10.2026 · 165 (15)
DOI: 10.1063/5.0352886 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 20.10.2026. Verfasst von Alexander L M Knoll und 13 weiteren Autor:innen. Beteiligte Einrichtung: Lehrstuhl für Theoretische Chemie II, Ruhr-Universität Bochum, 44780 Bochum, Germany. Die Arbeit nennt 75 Literaturangaben. Gefördert von: Deutsche Forschungsgemeinschaft, Studienstiftung des Deutschen Volkes.
Abstract (Originalsprache)
We present RuNNer 2.0, the “Ruhr University Neural Network energy representation,” a highly optimized software suite for training and evaluating high-dimensional neural network potentials (HDNNPs) of the second, third, and fourth generations. Long-range electrostatics and charge equilibration (QEq) for the description of non-local charge transfer in fourth-generation (4G) HDNNPs are accelerated by quasi-linear-scaling plane wave methods, reducing QEq computational complexity from O(N3) to O(Nlog2N) such that linear or quasi-linear scaling is achieved across all HDNNP generations. An optimized memory management strategy eliminates the training overhead traditionally associated with long-range interactions, allowing 4G-HDNNPs to be trained with the same efficiency as their local counterparts. Developed in modern Fortran (2003/2008 standards), combined with a hybrid Message Passing Interface (MPI)/Open Multi-Processing (OpenMP) parallelization scheme, RuNNer 2.0 has been designed to run efficiently in any central processing unit (CPU) environment, from cost-effective local workstations to massive high-performance computing (HPC) clusters. Its modular library architecture facilitates straightforward binding to external simulation software; native interfaces to the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) and the atomic simulation environment (ASE) provide full access to all its features, including built-in committee-based uncertainty quantification. The high efficiency and scalability of the RuNNer 2.0 ecosystem are demonstrated through detailed benchmarks.
Autor:innen und Institutionen
- Lehrstuhl für Theoretische Chemie II, Ruhr-Universität Bochum, 44780 Bochum, Germany.
- Research Center Chemical Sciences and Sustainability, Research Alliance Ruhr, 44780 Bochum, Germany.
- University of Basel, Department of Physics, Klingelbergstrasse 82, CH-4056 Basel, Switzerland.
- PSI Center for Scientific Computing, Theory and Data, Paul Scherrer Institute, 5232 Villigen PSI, Switzerland.
- ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 2, 8093 Zurich, Switzerland.
ORCID-Profile: Alexander L M Knoll, Moritz R Schäfer, K Nikolas Lausch, Moritz Gubler, Henry Wang, Richard Springborn, Redouan El Haouari, Alea Miako Liebetrau, Jonas A Finkler, Emir Kocer, Marco Eckhoff, Stefan Goedecker, Gunnar Schmitz, Jörg Behler
Förderung
- Deutsche Forschungsgemeinschaft
- Studienstiftung des Deutschen Volkes
Zitieren
APA
Knoll, A. L. M., Schäfer, M. R., Lausch, K. N., Gubler, M., Wang, H., Springborn, R., El Haouari, R., Liebetrau, A. M., Finkler, J. A., Kocer, E., Eckhoff, M., Goedecker, S., Schmitz, G., & Behler, J. (2026). RuNNer 2.0: A software suite for high-dimensional neural network potentials. The Journal of Chemical Physics, 165(15). https://doi.org/10.1063/5.0352886
Vancouver
Knoll ALM, Schäfer MR, Lausch KN, Gubler M, Wang H, Springborn R, et al. RuNNer 2.0: A software suite for high-dimensional neural network potentials. The Journal of Chemical Physics. 2026;165(15). doi:10.1063/5.0352886
BibTeX
@article{Knoll2026,
author = {Knoll, Alexander L M and Schäfer, Moritz R and Lausch, K Nikolas and Gubler, Moritz and Wang, Henry and Springborn, Richard and El Haouari, Redouan and Liebetrau, Alea Miako and Finkler, Jonas A and Kocer, Emir and Eckhoff, Marco and Goedecker, Stefan and Schmitz, Gunnar and Behler, Jörg},
title = {RuNNer 2.0: A software suite for high-dimensional neural network potentials},
journal = {The Journal of Chemical Physics},
year = {2026},
volume = {165},
number = {15},
doi = {10.1063/5.0352886}
}
EUVIMED