用费舍信息指导模型重组，自动设计更精准的原子间势能模型。
Composable and adaptive design of machine learning interatomic potentials guided by Fisher-information analysis
- 基于费舍信息矩阵动态组合单体模型并迭代优化。
- 在铌数据集上实现0.172 eV/Å力误差和0.013 eV/原子能量误差。
- 适合需要高精度且可扩展的材料模拟研究者使用。
本文提出一种自适应的物理启发式机器学习原子间势能（MLIP）建模策略。该策略通过从单项模型中迭代重构复合模型，并配合统一训练流程实现优化。同时，提出基于费舍信息矩阵（FIM）与多属性误差指标的模型评估方法，用于指导模型重构与超参数调优。结合重构与评估模块，构建了一种兼顾灵活性与可扩展性的自适应MLIP设计框架。以结构多样化的铌数据集为例，成功获得由75个参数构成的最优模型配置，在力预测上达到0.172 eV/Å的均方根误差，能量预测误差为0.013 eV/原子。
原文摘要 · Abstract (English)
An adaptive physics-inspired model design strategy for machine-learning interatomic potentials (MLIPs) is proposed. This strategy relies on iterative reconfigurations of composite models from single-term models, followed by a unified training procedure. A model evaluation method based on the Fisher information matrix (FIM) and multiple-property error metrics is also proposed to guide the model reconfiguration and hyperparameter optimization. By combining the reconfiguration and the evaluation subroutines, we provide an adaptive MLIP design strategy that balances flexibility and extensibility. In a case study of designing models against a structurally diverse niobium dataset, we managed to obtain an optimal model configuration with 75 parameters generated by our framework that achieved a force RMSE of 0.172 eV/Å and an energy RMSE of 0.013 eV/atom.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。