arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.22367physics.chem-phphysics.atm-clus
收藏

对分子红外光谱的机器学习互原子势的基准测试

Benchmarking machine-learned interatomic potentials for molecular infrared spectroscopy

Technical University of Munich(慕尼黑工业大学);Aalto University(阿尔托大学);Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心);Munich Data Science Institute(慕尼黑数据科学研究所);Department of Chemistry and Material Science, Aalto University(阿尔托大学化学与材料科学系)

Nitik Bhatia, Ondrej Krejci, Patrick Rinke

AI总结:

本文研究了五种消息传递神经网络架构在预测小有机分子红外光谱中的性能,评估了它们在计算效率、准确性和鲁棒性方面的表现,并比较了不同模型在转移能力和谱准确性上的差异。

摘要:

机器学习已通过使开发出的互原子势在计算上高效且高度准确而改变了原子模拟领域。这些进步使分子振动建模和预测红外光谱成为可能,其精度接近从头计算,但计算成本仅为一小部分。在这些方法中,消息传递神经网络(MPNN)已成为表示复杂原子相互作用的特别强大的模型类别。在本研究中,我们对五种MPNN架构(SchNet、FieldSchNet、SO3Net、PaiNN和MACE)进行基准测试,以预测小有机分子的红外光谱。SchNet和FieldSchNet是不变模型,而SO3Net、PaiNN和MACE是等变的,明确考虑了分子表示中的旋转对称性。我们评估了它们在计算效率、准确性和鲁棒性方面的性能。所有模型都能准确预测用于红外光谱计算所需的性质,如能量、力和偶极矩。它们还能够以高保真度捕捉分子动力学中分子的谐振频率和红外光谱。然而,SchNet和FieldSchNet在未见系统中转移性有限,而SO3Net、PaiNN和MACE则更具泛化能力。在计算效率方面,SchNet最高效,FieldSchNet能够实现场依赖响应建模但成本更高。PaiNN在准确性和效率之间取得了最佳平衡,MACE提供了最高谱准确性和转移性,而SO3Net的性能介于PaiNN和MACE之间。

英文摘要:

Machine learning has transformed the field of atomistic simulations by enabling the development of interatomic potentials that are computationally efficient and highly accurate. These advances have opened the door to modeling molecular vibrations and predicting infrared spectra with near ab-initio accuracy at a fraction of the computational cost. Among these approaches, message-passing neural networks (MPNNs) have emerged as a particularly powerful class of models for representing complex atomic interactions. In this study, we benchmark five MPNN architectures, SchNet, FieldSchNet, SO3Net, PaiNN, and MACE, for predicting infrared spectra of small organic molecules. SchNet and FieldSchNet are invariant models, while SO3Net, PaiNN, and MACE are equivariant, explicitly accounting for rotational symmetries in molecular representations. We evaluate their performance in terms of computational efficiency, accuracy, and robustness. All models accurately predict properties, such as energies, forces, and dipole moments, required for infrared spectra calculations. They also capture harmonic frequencies and infrared spectra derived from molecular dynamics with high fidelity for molecules in the training set. However, SchNet and FieldSchNet show limited transferability to unseen systems, while SO3Net, PaiNN, and MACE generalize more effectively. In terms of computational efficiency, SchNet is the most efficient and FieldSchNet enables field-dependent response modeling but with higher cost. PaiNN achieves the best balance between accuracy and efficiency, MACE provides the highest spectral accuracy and transferability, and SO3Net performs between PaiNN and MACE.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑