关于DM21神经网络DFT函数用于化学计算的实际应用:专注于几何优化
Kirill Kulaev1, Alexander Ryabov1, Michael G Medvedev2
1Skolkovo Institute of Science and Technology, Artificial Intelligence Center, Moscow 121205, Russian Federation.
像DM21这样的神经网络功能引入了数值噪声,阻碍了精确的分子几何优化. 这项研究表明,它们比量子化学计算的传统方法更慢,更不精确.
科学领域:
- 量子化学 是一个量子化学.
- 计算化学计算化学
背景情况:
- 密度函数理论 (DFT) 是量子化学的基石,它平衡了精度和计算速度.
- 基于神经网络的交换关联 (XC) 函数,如谷歌DeepMind的DM21,是新兴的替代方案.
研究的目的:
- 为了评估DM21功能用于分子几何优化的效率.
- 研究神经网络衍生的XC能量和潜在的非平滑性对几何精度的影响.
主要方法:
- 使用PySCF.中实现的DM21函数的几何优化.
- 在基准数据集上对DM21性能与传统XC函数的比较.
- 分析神经网络输出中的数值噪声及其对核梯度的影响.
主要成果:
- 来自DM21输出的数字噪声污染了核梯度,这对于几何优化至关重要.
- 对于光滑的梯度,需要0.00010.001 Å的数值分化步骤.
- 通过将噪声添加到分析性SCAN功能能量中,可以模拟DM21的非光滑性.
结论:
- 在优化分子几何学准确度方面,DM21不超过分析函数.
- DM21比传统的函数显著慢,这限制了它在量子化学中的实际应用.
- 神经网络函数的非平滑性对精确的计算化学应用提出了挑战.
更多相关视频
08:54Vibrational Spectra of a N719-Chromophore/Titania Interface from Empirical-Potential Molecular-Dynamics Simulation, Solvated by a Room Temperature Ionic Liquid
Published on: January 25, 2020
05:57Author Spotlight: In Silico Creation and Impact of Carbonylated Amino Acids on Protein Structure and Function
Published on: April 26, 2024
相关概念视频
Molecular Geometry and Dipole Moments
Predicting Molecular Geometry
UV–Vis Spectroscopy: Woodward–Fieser Rules
Valence Bond Theory and Hybridized Orbitals
A σ bond (single bond in a Lewis structure) is a covalent bond in which the electron density is...
Crystal Field Theory - Octahedral Complexes
To explain the observed behavior of transition metal complexes (such as colors), a model involving electrostatic interactions between the electrons from the ligands and the electrons in the unhybridized d orbitals of the central metal atom has been developed. This electrostatic model is crystal field theory (CFT). It helps to understand, interpret, and predict the colors, magnetic behavior, and some structures of coordination compounds of transition metals.
CFT focuses on...
¹³C NMR: Distortionless Enhancement by Polarization Transfer (DEPT)
