Gaussian approximation potentials: Theory, software implementation and application examples.

Gaussian approximation potentials: Theory, software implementation and application examples.
复制标题

DOI:
10.1063/5.0160898
复制
发表时间:
2023-10
期刊:
The Journal of chemical physics
影响因子:
--
通讯作者:
Sascha Klawohn;James P Darby;J. Kermode;Gábor Csányi;Miguel A Caro;Albert P. Bart'ok
Sascha Klawohn;James P Darby;J. Kermode;Gábor Csányi;Miguel A Caro;Albert P. Bart'ok
中科院分区:
其他
文献类型:
--
作者:
Sascha Klawohn;James P Darby;J. Kermode;Gábor Csányi;Miguel A Caro;Albert P. Bart'ok

文献摘要

相似文献

高斯近似势(Gaussian Approximation Potential,GAP)是一类机器学习的原子间势,通常用于在原子尺度上对材料和分子系统进行建模。软件实现提供了两个拟合模型使用从头算数据和使用原子模拟中产生的潜力的手段。详细的GAP理论,算法和软件,连同详细的使用示例,以帮助新的和现有的用户。我们回顾了差距框架的一些最新进展,包括消息传递接口的拟合代码的并行化,使其能够在数千个中央处理器核心上使用,以及描述符的压缩,以消除不同化学元素数量的不良缩放。
Gaussian Approximation Potentials (GAPs) are a class of Machine Learned Interatomic Potentials routinely used to model materials and molecular systems on the atomic scale. The software implementation provides the means for both fitting models using ab initio data and using the resulting potentials in atomic simulations. Details of the GAP theory, algorithms and software are presented, together with detailed usage examples to help new and existing users. We review some recent developments to the GAP framework, including Message Passing Interface parallelisation of the fitting code enabling its use on thousands of central processing unit cores and compression of descriptors to eliminate the poor scaling with the number of different chemical elements.