arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2401.05223cond-mat.mtrl-sci
收藏

基于结构的材料性能预测中物理引导的双自监督学习

Physics guided dual Self-supervised learning for structure-based materials property prediction

Nihang Fu, Lai Wei, Jianjun Hu

AI总结:

针对材料性能预测中标注数据稀缺的问题,提出物理引导的双自监督学习框架DSSL,结合预测性与对比性SSL策略并引入微观性能预测作为前置任务,显著提升了GNN模型的预测性能。

摘要:

深度学习(DL)模型现已广泛应用于高性能材料性能预测,例如形成能和带隙。然而,训练此类DL模型通常需要大量标注数据,而对于大多数材料性能(如剥离能和弹性能)而言,通常无法获得这些数据。自监督学习(SSL)方法被提出以解决这一数据稀缺问题,它通过在各个研究领域中从无标签数据学习内在表示来实现。在此,我们提出了DSSL,一个物理引导的双SSL框架,用于基于图神经网络(GNNs)的材料性能预测。该混合框架将基于节点掩码的预测性SSL与基于原子坐标扰动的对比性SSL策略相结合,使其能够学习捕获输入晶体局部和全局信息的结构嵌入。特别是,我们提出将预测与宏观性能(如弹性)相关的微观性能(如原子刚度)作为额外的前置任务,以实现物理引导的预训练过程。我们在Materials Project数据库上使用无标签数据预训练DSSL模型,并在十个具有不同材料性能的额外数据集上进行微调。实验结果表明,使用SSL策略教授神经网络一些物理知识,与基线GNN模型相比,可带来高达26.89%的性能提升。我们的源代码现已在 https://github.com/usccolumbia/DSSL 免费提供。

英文摘要:

Deep learning (DL) models have now been widely used for high-performance material property prediction for properties such as formation energy and band gap. However, training such DL models usually requires a large amount of labeled data, which is usually not available for most materials properties such as exfoliation energy and elastic properties. Self-supervised learning (SSL) methods have been proposed to address this data scarcity issue by learning inherent representations from unlabeled data in various research fields. Herein, we present DSSL, a physics-guided Dual SSL framework, for graph neural networks (GNNs) based material property prediction. This hybrid framework combines node-masking based predictive SSL with atomic coordinate perturbation based contrastive SSL strategies, allowing it to learn structural embeddings that capture both local and global information of input crystals. Especially, we propose to use predicting the macroproperty (e.g. elasticity) related microproperty such as atomic stiffness as an additional pretext task to achieve physics-guided pretraining process. We pretrain our DSSL model on the Materials Project database with unlabeled data and finetune it with ten extra datasets with different material properties. The experimental results demonstrate that teaching neural networks some physics using the SSL strategy can bring up to 26.89\% performance improvement compared to the baseline GNN models. Our source code is now freely available at https://github.com/usccolumbia/DSSL

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。

↑