arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.08032cond-mat.mtrl-scics.LG
收藏

基于结构的分布外(OOD)材料性质预测:一项基准研究

Structure-based out-of-distribution (OOD) materials property prediction: a benchmark study

University of South Carolina(南卡罗来纳大学)

Sadman Sadeed Omee, Nihang Fu, Rongzhi Dong, Ming Hu, Jianjun Hu

AI总结:

本文通过构建五类分布外材料性质预测基准,系统评估了多种图神经网络,发现现有最先进模型在OOD任务上存在显著泛化差距,并分析了CGCNN等模型更稳健的原因。

摘要:

在现实世界的材料研究中,机器学习(ML)模型通常被期望能够预测和发现偏离已知材料的新型特殊材料。因此,如何对ML模型在分布外(OOD)材料性质预测中的表现进行客观评估,成为一个紧迫的问题,这些OOD材料不同于训练集的分布。由于典型材料数据集固有的冗余性,通过随机划分数据集来对材料性质预测模型进行传统性能评估,常常会导致人为偏高的性能评价。在此,我们提出了一项针对基于结构的图神经网络(GNNs)在外推式OOD材料性质预测方面的全面基准研究。我们针对来自MatBench研究的三个基准数据集,构建了五类不同的OOD ML问题。我们的大量实验表明,与MatBench研究中的基线相比,当前最先进的GNN算法在OOD性质预测任务上的平均表现显著不佳,这揭示了在现实材料预测任务中存在的关键泛化差距。我们进一步考察了这些GNN模型的潜在物理空间,并识别出CGCNN、ALIGNN和DeeperGATGNN的OOD性能显著优于MatBench研究中当前最佳模型(coGN和coNGN)的原因,并提供了改进其性能的见解。

英文摘要:

In real-world material research, machine learning (ML) models are usually expected to predict and discover novel exceptional materials that deviate from the known materials. It is thus a pressing question to provide an objective evaluation of ML model performances in property prediction of out-of-distribution (OOD) materials that are different from the training set distribution. Traditional performance evaluation of materials property prediction models through random splitting of the dataset frequently results in artificially high performance assessments due to the inherent redundancy of typical material datasets. Here we present a comprehensive benchmark study of structure-based graph neural networks (GNNs) for extrapolative OOD materials property prediction. We formulate five different categories of OOD ML problems for three benchmark datasets from the MatBench study. Our extensive experiments show that current state-of-the-art GNN algorithms significantly underperform for the OOD property prediction tasks on average compared to their baselines in the MatBench study, demonstrating a crucial generalization gap in realistic material prediction tasks. We further examine the latent physical spaces of these GNN models and identify the sources of CGCNN, ALIGNN, and DeeperGATGNN's significantly more robust OOD performance than those of the current best models in the MatBench study (coGN and coNGN), and provide insights to improve their performance.

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。

↑