arXiv：cs.LG· Hongwei Du, Dingyang Lv, Baole Wei, Yu Ren, Feng Yu, Xin He, Bonan Zhu, Yongda Huang, Yongheng Li, Jianjun Liu, Siqi Shi, Hong Wang, Ziheng Lu·· 4 小时前AI 评分34
通用机器学习力场在多组分材料中的误差来源研究
Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials
AI 导读
研究构建了7,599个多组分构型基准，评估11个预训练机器学习力场模型对密度泛函理论的能量、力和应力预测。压缩侧力误差是拉伸侧的1.81-1.95倍，训练参考覆盖差异与误差增大存在定性关联。在参数匹配下，最大阶数为2和4的球谐表示分别降低了43种元素中38种和40种的测试力误差。
正文
Authors:Hongwei Du, Dingyang Lv, Baole Wei, Yu Ren, Feng Yu, Xin He, Bonan Zhu, Yongda Huang, Yongheng Li, Jianjun Liu, Siqi Shi, Hong Wang, Ziheng Lu
Abstract:Universal machine learning force-field generalization to multicomponent environments generated by compositional design remains insufficiently assessed. We construct a benchmark of 7,599 multicomponent configurations inspired by high-entropy design, elemental substitution and anion mixing. Eleven pretrained models are evaluated against density functional theory for energies, forces and stresses, with assessment extended to elastic, vibrational and adsorption-related properties. Force errors are analysed through training-reference coverage, local geometric heterogeneity, distance directionality and elemental response. Distances to training-reference environments reveal a qualitative association between coverage differences and increasing errors, while substantial variation remains at similar distances. Higher-error groups show greater local geometric heterogeneity, although OMat24 provides broad coverage of these environments. Relative to training-reference pair medians, errors remain low near the median, rise steeply on the compression side and increase more weakly on the extension side. After matching element pairs and absolute distance deviations, compression-side force errors are 1.81-1.95 times extension-side errors. Model-predicted pairwise interaction curves show greater curvature under compression. Fitting difficulty in independent elemental systems correlates with electronic band-energy responses to atomic displacements and Fermi-level shifts, and a similar pattern is observed in multicomponent systems. In parameter-matched comparisons, spherical-harmonic representations with maximum degrees of 2 and 4 lower test force errors for 38 and 40 of 43 elements, respectively, while differences in elemental difficulty remain. These findings inform force-field selection for experimental compositional design and identify targets for training-data sampling and model representations.
| Comments: | 20 pages, 10 figures |
| Subjects: | Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG); Computational Physics (physics.comp-ph) |
| Cite as: | arXiv:2610.09837 [cond-mat.mtrl-sci] |
| (or arXiv:2610.09837v1 [cond-mat.mtrl-sci] for this version) | |
| https://doi.org/10.48550/arXiv.2610.09837 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: HongWei Du [view email]
[v1]
Wed, 7 Oct 2026 11:01:36 UTC (4,232 KB)
来源：arXiv：cs.LG · arxiv.org