GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations
Zheyong Fan, Yanzhou Wang, Penghua Ying, Keke Song, Junjie Wang, Yong Wang, Zezhu Zeng, Ke Xu, Eric Lindgren, J. Magnus Rahm, Alexander J. Gabourie, Jiahui Liu,… 展开作者
Zheyong Fan, Yanzhou Wang, Penghua Ying, Keke Song, Junjie Wang, Yong Wang, Zezhu Zeng, Ke Xu, Eric Lindgren, J. Magnus Rahm, Alexander J. Gabourie, Jiahui Liu, Haikuan Dong, Jianyang Wu, Yue Chen, Zheng Zhong, Jian Sun, Paul Erhart, Yanjing Su, Tapio Ala-Nissila
我们展示了基于[Fan et al., Phys. Rev. B 104, 104309 (2021)]中引入的neuroevolution potential（NEP）框架的机器学习势（MLPs）的最新进展，以及它们在开源软件包GPUMD中的实现。我们通过两种方式提高了NEP模型的精度：一是使用Chebyshev基函数的线性组合改进原子环境描述符中的径向函数，二是像atomic cluster expansion方法那样，用一些四体和五体贡献扩展角向描述符。我们还详细介绍了NEP方法在图形处理单元（GPU）中的高效实现以及构建NEP模型的工作流程，并展示了它们在大规模原子模拟中的应用。通过与最先进的MLPs进行比较，我们表明NEP方法不仅达到了高于平均水平的精度，而且计算效率远高于现有方法。这些结果表明，GPUMD软件包是解决需要高精度、大规模原子模拟的挑战性问题的有前景的工具。为了能够使用最小的训练集构建MLPs，我们提出了一种基于预训练NEP模型潜在空间的主动学习方案。最后，我们介绍了三个独立的Python软件包：GPYUMD、CALORINE和PYNEP，它们使GPUMD能够集成到Python工作流程中。
英文摘要：
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 contributions as in the atomic cluster expansion approach. We also detail our efficient implementation of the NEP approach in graphics processing units as well as our workflow for the construction of NEP models, and we demonstrate their application in large-scale atomistic simulations. By comparing to state-of-the-art MLPs, we show that the NEP approach not only achieves above-average accuracy but also is far more computationally efficient. These results demonstrate that the GPUMD package is a promising tool for solving challenging problems requiring highly accurate, large-scale atomistic simulations. To enable the construction of MLPs using a minimal training set, we propose an active-learning scheme based on the latent space of a pre-trained NEP model. Finally, we introduce three separate Python packages, GPYUMD, CALORINE, and PYNEP, which enable the integration of GPUMD into Python workflows.
Neuroevolution machine learning potentials: Combining high accuracy and low cost in atomistic simulations and application to heat transport
我们开发了一个神经进化潜力（NEP）框架，用于生成基于神经网络的机器学习潜力。他们使用进化策略进行大规模分子动力学（MD）模拟训练。基于切比雪夫多项式和勒让德多项式构造了原子环境的描述子。该方法在开源GPUMD包内的图形处理单元中实现，使用一个Nvidia Tesla V100可以获得每秒超过$10^7$原子步的计算速度。此外，每个原子的热流是在NEP，这铺平了道路，有效和准确的MD模拟材料的热输运与强声子非谐性或空间无序，这通常不能准确地处理无论是传统的经验势或微扰方法。
qNEP: A highly efficient neuroevolution potential with dynamic charges for large-scale atomistic simulations
College of Physical Science and Technology, Bohai University（渤海大学物理科学与技术学院）；Suzhou Laboratory（苏州实验室）；Chalmers University of Technology（查尔姆斯理工大学）；Department of Materials Science and Engineering, Westlake University（西湖大学材料科学与工程系）；School of Science, Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳）理学院）；Shenzhen Key Laboratory of Micro/Nano-Porous Functional Materials (SKLPM), Department of Materials Science and Engineering, Southern University of Science and Technology（南方科技大学材料科学与工程系微/纳孔功能材料深圳市重点实验室）；Department of Materials Science and Engineering, City University of Hong Kong（香港城市大学材料科学与工程系）
NEPMaker: Active learning of neuroevolution machine learning potential for large cells
National Laboratory of Solid State Microstructures, School of Physics and Collaborative Innovation Center of Advanced Microstructures, Nanjing University（南京大学固体微结构物理学院、前沿科学中心、先进微结构协同创新中心）；College of Physical Science and Technology, Bohai University（渤海大学物理科学与技术学院）
University of Zurich（苏黎世大学）；University of Regensburg（雷根斯堡大学）；Helmholtz Zentrum Dresden-Rossendorf（德累斯顿-罗斯多夫亥姆霍兹中心）；ETH Zurich（苏黎世联邦理工学院）；Paderborn University（帕德博恩大学）；Christian-Albrechts-University Kiel（基尔大学）；Cambridge University（剑桥大学）；McGill University（麦吉尔大学）；HQS Quantum Simulations GmbH（HQS量子模拟有限公司）
Coupling all-electron full-potential density functional theory with grid-based continuum embeddings
Boise State University（博伊西州立大学）；PSI Center for Scientific Computing, Theory and Data（保罗谢勒研究所科学计算、理论与数据中心）；Fritz Haber Institute of the Max Planck Society（马克斯·普朗克学会弗里茨·哈伯研究所）
GPU-MetaD: Full-Life-Cycle GPU Accelerated Metadynamics with Machine Learning Potentials
National Laboratory of Solid State Microstructures, School of Physics and Collaborative Innovation Center of Advanced Microstructures, Nanjing University（南京大学固体微结构物理国家重点实验室、物理学院和先进微结构协同创新中心）；College of Physical Science and Technology, Bohai University（渤海大学物理科学与技术学院）
