机器学习原子间势中长程静电相互作用的通用增强框架
A universal augmentation framework for long-range electrostatics in machine learning interatomic potentials
- Department of Chemistry, UC Berkeley, California 94720, United States（化学系，伯克利大学，加利福尼亚州，94720，美国）
- Bakar Institute of Digital Materials for the Planet, UC Berkeley, California 94720, United States（地球数字材料研究所，伯克利大学，加利福尼亚州，94720，美国）
- The Institute of Science and Technology Austria, Am Campus 1, 3400 Klosterneuburg, Austria（奥地利科学与技术研究所）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出通用长程静电增强框架LES，可兼容多种短程机器学习原子间势，无需电学性质标签即可推断静电、极化与Born有效电荷，并在多个体系及SPICE数据集上验证其提升精度和泛化能力。
AI中文摘要：
当前大多数机器学习原子间势（MLIPs）依赖短程近似，未显式处理长程静电相互作用。为解决这一问题，我们近期开发了隐式Ewald求和（Latent Ewald Summation, LES）方法，该方法仅通过从能量和力的训练数据中学习，即可推断静电相互作用、极化以及Born有效电荷（BECs）。在此，我们将LES作为一个独立库呈现，兼容任何短程MLIP，并展示了其与MACE、NequIP、CACE和CHGNet等方法的集成。我们在不同体系上对LES增强模型进行了基准测试，包括体相水、极性二肽以及缺陷基底上的金二聚体吸附，结果表明LES不仅能正确捕捉静电效应，还能提高精度。此外，我们通过在包含分子和团簇的SPICE数据集上训练MACELES-OFF，将LES扩展到大规模且化学多样性的数据，从而构建了一个适用于包括生物分子在内的有机体系的具有静电相互作用的通用MLIP。MACELES-OFF比在相同数据集上训练的短程对应模型（MACE-OFF）更准确，能够可靠地预测偶极矩和BECs，并且对体相液体的描述更优。通过在不直接训练电学性质的情况下实现高效的长程静电相互作用，LES为静电基础型MLIPs铺平了道路。
英文摘要：
Most current machine learning interatomic potentials (MLIPs) rely on short-range approximations, without explicit treatment of long-range electrostatics. To address this, we recently developed the Latent Ewald Summation (LES) method, which infers electrostatic interactions, polarization, and Born effective charges (BECs), just by learning from energy and force training data. Here, we present LES as a standalone library, compatible with any short-range MLIP, and demonstrate its integration with methods such as MACE, NequIP, CACE, and CHGNet. We benchmark LES-enhanced models on distinct systems, including bulk water, polar dipeptides, and gold dimer adsorption on defective substrates, and show that LES not only captures correct electrostatics but also improves accuracy. Additionally, we scale LES to large and chemically diverse data by training MACELES-OFF on the SPICE set containing molecules and clusters, making a universal MLIP with electrostatics for organic systems including biomolecules. MACELES-OFF is more accurate than its short-range counterpart (MACE-OFF) trained on the same dataset, predicts dipoles and BECs reliably, and has better descriptions of bulk liquids. By enabling efficient long-range electrostatics without directly training on electrical properties, LES paves the way for electrostatic foundation MLIPs.
