GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations.

  title={GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations.},
  author={Zheyong Fan and Yanzhou Wang and Penghua Ying and Keke Song and Junjie Wang and Yong Wang and Zezhu Zeng and Ke Xu and Eric B. Lindgren and J. Magnus Rahm and Alexander J. Gabourie and Jiahui Liu and Haikuan Dong and Jianyang Wu and Yue Chen and Zheng Zhong and Jian Sun and Paul Erhart and Yan Jing Su and Tapio Ala‐Nissila},
  journal={The Journal of chemical physics},
  volume={157 11},
We present our latest advancements of machine-learned potentials (MLPs) based on the neuroevolution potential (NEP) framework introduced in Fan et al. [Phys. Rev. B 104, 104309 (2021)] and their implementation in the open-source package gpumd. We increase the accuracy of NEP models both by improving the radial functions in the atomic-environment descriptor using a linear combination of Chebyshev basis functions and by extending the angular descriptor with some four-body and five-body… 

Quantum-corrected thickness-dependent thermal conductivity in amorphous silicon predicted by machine-learning molecular dynamics simulations

Amorphous silicon (a-Si) is an important thermal-management material and also serves as an ideal playground for studying heat transport in strongly disordered materials. Theoretical prediction of the

Machine learning based modeling of disordered elemental semiconductors: understanding the atomic structure of a-Si and a-C

Disordered elemental semiconductors, most notably a-C and a-Si, are ubiquitous in a myriad of different applications. These exploit their unique mechanical and electronic properties. In the past

Variable thermal transport in black, blue, and violet phosphorene from extensive atomistic simulations with a neuroevolution potential

Phosphorus has diverse chemical bonds and even in its two-dimensional form there are three stable allotropes: black phosphorene (Black-P), blue phosphorene (Blue-P), and violet phosphorene

Anisotropic and high thermal conductivity in monolayer quasi-hexagonal fullerene: A comparative study against bulk phase fullerene

Recently a novel two-dimensional (2D) C 60 based crystal called quasi-hexagonal-phase fullerene (QHPF) has been fabricated and demonstrated to be a promising candidate for 2D electronic devices [Hou



PyXtal_FF: a python library for automated force field generation

The aim of PyXtal_FF is to promote the application of atomistic simulations through providing several choices of atom-centered descriptors and machine learning regressions in one platform, which can train MLPs with either generalized linear regression or neural network models.

Perspective: Machine learning potentials for atomistic simulations.

  • J. Behler
  • Materials Science
    The Journal of chemical physics
  • 2016
Recent advances in machine learning (ML) now offer an alternative approach for the representation of potential-energy surfaces by fitting large data sets from electronic structure calculations, which are reviewed along with a discussion of their current applicability and limitations.

Pushing the Limit of Molecular Dynamics with Ab Initio Accuracy to 100 Million Atoms with Machine Learning

  • Weile JiaHan Wang Linfeng Zhang
  • Computer Science
    SC20: International Conference for High Performance Computing, Networking, Storage and Analysis
  • 2020
A machine learningbased simulation protocol (Deep Potential Molecular Dynamics), while retaining ab initio accuracy, can simulate more than 1 nanosecond-long trajectory of over 100 million atoms per day, using a highly optimized code (GPU DeePMD-kit) on the Summit supercomputer.

Heterogeneous Parallelization and Acceleration of Molecular Dynamics Simulations in GROMACS

This work presents the heterogeneous parallelization and acceleration design of molecular dynamics implemented in the GROMACS codebase over the last decade, and enables excellent performance from single GPU simulations through strong scaling across multiple GPUs and efficient multi-node parallelization.

Strategies for the construction of machine-learning potentials for accurate and efficient atomic-scale simulations

This tutorial review aims at equipping computational chemists and materials scientists with the required background knowledge for ANN potential construction and application, with the intention to accelerate the adoption of the method so that it can facilitate exciting research that would otherwise be challenging with conventional strategies.

Optimizing many-body atomic descriptors for enhanced computational performance of machine learning based interatomic potentials

  • M. Caro
  • Physics
    Physical Review B
  • 2019
We explore different ways to simplify the evaluation of the smooth overlap of atomic positions (SOAP) many-body atomic descriptor [Bart\'ok et al., Phys. Rev. B 87, 184115 (2013).]. Our aim is to

FCHL revisited: Faster and more accurate quantum machine learning.

The revised FCHL19 representation is introduced, a revision of the previous work where the representation is discretized and the individual features are rigorously optimized using Monte Carlo optimization, that is fast and lightweight enough for use in general chemistry problems as well as molecular dynamics simulations.