训练后可动态调整截断半径，提升机器学习势能模型的灵活性与效率。
Flexible Cutoff Learning: Optimizing Machine Learning Potentials After Training
- 训练时随机采样原子截断半径，使模型具备可调性
- 优化后计算成本降低60%以上，力误差增加不足1%
- 无需重训即可适配不同应用场景，适合高效分子模拟
我们提出灵活截断学习（FCL），一种可在训练后调整截断半径的机器学习原子间势能（MLIP）方法。与传统固定截断半径不同，FCL在训练中为每个原子独立随机采样截断半径。训练完成后，可根据具体应用设置不同的原子级截断半径，实现精度-成本权衡的定制化优化。结合可微分代价模型，可在训练后对特定体系优化每原子截断半径。我们在MAD数据集上使用改进的MACE架构验证了FCL，在包含分子晶体的子集上，优化后的截断半径使计算成本降低超过60%，同时力误差增加小于1%。结果表明，FCL可训练出单一通用型MLIP，通过训练后截断优化适配多种应用，避免重复训练。
原文摘要 · Abstract (English)
We introduce Flexible Cutoff Learning (FCL), a method for training machine learning interatomic potentials (MLIPs) whose cutoff radii can be adjusted after training. Unlike conventional MLIPs that fix the cutoff radius during training, FCL models are trained by randomly sampling cutoff radii independently for each atom. The resulting model can then be deployed with different per-atom cutoff radii depending on the application, enabling application-specific optimization of the accuracy-cost tradeoff. Using a differentiable cost model, these per-atom cutoffs can be optimized for specific target systems after training. We demonstrate FCL with a modified MACE architecture trained on the MAD dataset. For a subset featuring molecular crystals, optimized per-atom cutoffs reduce computational cost by more than 60% while increasing force errors by less than 1%. These results show that FCL enables training of a single general-purpose MLIP that can be adapted to diverse applications through post-training cutoff optimization, eliminating the need for retraining.
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。