材料科学中的断裂神经缩放定律
Broken neural scaling laws in materials science
AI总结：
本研究发现材料科学中预测介电函数的神经缩放定律在数据集规模上失效，但模型参数数量的缩放遵循简单幂律。
摘要：
在材料科学中，数据稀缺且昂贵，无论是计算还是实验生成。因此，识别模型性能如何随数据集大小和模型容量变化至关重要，以区分数据受限和模型受限的 regime。神经缩放定律提供了一个框架来量化这种行为，并指导材料数据集和机器学习架构的设计。在此，我们研究了神经缩放定律用于一个典型的材料科学任务：预测金属的介电函数，这是一个高维响应，决定了固体如何与光相互作用。使用超过20万来自高通量ab initio计算的介电函数，我们研究了两个多目标图神经网络，训练以预测频率依赖的复能带介电函数和Drude频率。我们观察到相对于数据集大小的神经缩放定律被打破，而随着模型参数数量的缩放则遵循简单的幂律，迅速饱和。
英文摘要：
In materials science, data are scarce and expensive to generate, whether computationally or experimentally. Therefore, it is crucial to identify how model performance scales with dataset size and model capacity to distinguish between data- and model-limited regimes. Neural scaling laws provide a framework for quantifying this behavior and guide the design of materials datasets and machine learning architectures. Here, we investigate neural scaling laws for a paradigmatic materials science task: predicting the dielectric function of metals, a high-dimensional response that governs how solids interact with light. Using over 200,000 dielectric functions from high-throughput ab initio calculations, we study two multi-objective graph neural networks trained to predict the frequency-dependent complex interband dielectric function and the Drude frequency. We observe broken neural scaling laws with respect to dataset size, whereas scaling with the number of model parameters follows a simple power law that rapidly saturates.
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。
