arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.00239cond-mat.mtrl-sci
收藏

用于材料设计的可解释机器学习

Interpretable Machine Learning for Materials Design

James Dean, Matthias Scheffler, Thomas A. R. Purcell, Sergey V. Barabash, Rahul Bhowmik, Timur Bazhirov

摘要:

在广泛采用机器学习(ML)和高通量材料筛选的推动下,以数据为中心的材料设计方法已经成为一种强大而强大的工具,用于对材料性能进行计算机预测。当训练模型来预测材料属性时,研究人员经常面临模型的可解释性或其性能之间的困难选择。我们通过利用四种不同的最先进的ML技术来研究这种权衡:XGBoost,SISO,Roost和TPOT,用于预测钙钛矿和2D材料的结构和电子特性。然后,我们评估了ML继续整合到材料发现中的未来前景,并确定了随着文献数据集的大小和模型复杂性的增加,将继续挑战研究人员的关键问题。最后,我们为这些挑战提供了几种可能的解决方案,重点是保持可解释性,并分享我们对放大ML对材料设计的影响的想法。

英文摘要:

Fueled by the widespread adoption of Machine Learning (ML) and the high-throughput screening of materials, the data-centric approach to materials design has asserted itself as a robust and powerful tool for the in-silico prediction of materials properties. When training models to predict material properties, researchers often face a difficult choice between a model's interpretability or its performance. We study this trade-off by leveraging four different state-of-the-art ML techniques: XGBoost, SISSO, Roost, and TPOT for the prediction of structural and electronic properties of perovskites and 2D materials. We then assess the future outlook of the continued integration of ML into materials discovery and identify key problems that will continue to challenge researchers as the size of the literature's datasets and complexity of models increases. Finally, we offer several possible solutions to these challenges with a focus on retaining interpretability and share our thoughts on magnifying the impact of ML on materials design.

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。

↑