Machine-learning potentials are usually trained on the ground-state, Born-Oppenheimer energy surface, which depends exclusively on the atomic positions and not on the simulation temperature. This disregards the effect of thermally-excited electrons, that is important in metals, and essential to the description of warm dense matter. An accurate physical description of these effects requires that the nuclei move on a temperature-dependent electronic free energy. We propose a method to obtain machine-learning predictions of this free energy at an arbitrary electron temperature using exclusively training data from ground-state calculations, avoiding the need to train temperature-dependent potentials, and benchmark it on metallic liquid hydrogen at the conditions of the core of gas giants and brown dwarfs. This work demonstrates the advantages of hybrid schemes that use physical consideration to combine machine-learning predictions, providing a blueprint for the development of similar approaches that extend the reach of atomistic modelling by removing the barrier between physics and data-driven methodologies.
Comment on "Breaking the theoretical scaling limit for predicting quasi-particle energies: The stochastic GW approach", by Daniel Neuhauser et al., arXiv:1402.5035v1
Predicting large-supercell defect formation energies from machine-learning charge density models trained on small supercells
College of Integrated Circuits and Micro-Nano Electronics, and Key Laboratory of Computational Physical Sciences (MOE), Fudan University(复旦大学集成电路与微纳电子学院,计算物理科学教育部重点实验室)
Predicting conductivities of alkali borophosphate glasses based on site energy distributions derived from network former unit concentrations
对于网络玻璃中的离子输运,根据结构特性预测电导率是一个巨大的挑战。为此,有必要了解离子发生热激活跳跃运动的能量景观。对于碱金属硼磷酸盐玻璃,提出了一种统计力学方法来预测移动碱金属离子驻留点能量分布的基本特征。从形成玻璃状网络的化学单元得到相应的位置能量分布。一个跳跃模型的基础上的网站能量景观允许模拟的电导率活化能的变化与硼酸盐磷酸盐的混合比。在这里,我们完善和扩展这种一般的方法,以应付最小的本地激活障碍,并计算直流电导率,而不需要进行广泛的蒙特卡罗模拟。这种计算依赖于映射的多体离子动力学到网络的本地电导来自移动离子的基本跳跃率。将理论模拟应用于三个系列的碱性硼磷酸盐玻璃,其组成为$0.33$ Li $_2$ O $-0.67$ [ $x$ B $_2$ O $_3$ - $(1\!-\!x)$ P $_2$ O $_5$ ],$0.35$ Na $_2$ O $-0.65$ [ $x$ B $_2$ O $_3$ - $(1\!-\!x)$ P $_2$ O $_5$ ]和$0.4$ Na $_2$ O $-0.6$ [ $x$ B $_2$ O $_3$ -$(1\!-\!x)$ P $_2$ O $_5$ ]与实验数据吻合良好。