Auto-MatRegressor材料性能自动预测器:解放材料机器学习"调参师"

Auto-MatRegressor:liberating machine learning alchemists

机器学习因其能够快速、精准拟合数据的潜在模式而被广泛应用于材料构效关系研究。然而,材料科学家往往需要进行繁琐的模型选择及参数寻优才能构建出高精度预测模型,为了解放材料机器学习"调参师",本文研发了基于元学习的材料性能自动预测器,采集了60份文献公开数据集与60份标准数据集,基于此训练18种常用回归算法并获得其预测性能,定义与计算了27个刻画数据集特点的元特征,以此构建了一份蕴含建模经验的元数据集;同时,创建了表征数据集所属材料类型的类别树,将其嵌入基于距离的元学习算法,进一步耦合贝叶斯优化算法,实现领域知识和元数据协同驱动下的自动算法推荐和模型参数确定,实验结果表明,材料科学家仅需为新材料性能预测任务提供数据集,便可利用该预测器高效地构建具有与文献报道相当或更高预测精度的机器学习模型.

Machine learning (ML) is widely used to uncover structure–property relationships of materials due to itsability to quickly find potential data patterns and make accurate predictions. However, like alchemists,materials scientists are plagued by time-consuming and labor-intensive experiments to build highaccuracy ML models. Here, we propose an automatic modeling method based on meta-learning for materials property prediction named Auto-MatRegressor, which automates algorithm selection and hyperparameter optimization by learning from previous modeling experience, i.e., meta-data on historicaldatasets. The meta-data used in this work consists of 27 meta-features that characterize the datasetsand the prediction performances of 18 algorithms commonly used in materials science. To recommendoptimal algorithms, a collaborative meta-learning method embedded with domain knowledge quantifiedby a materials categories tree is designed. Experiments on 60 datasets show that compared with the traditional modeling method from scratch, Auto-MatRegressor automatically selects appropriate algorithmsat lower computational cost, which accelerates constructing ML models with good prediction accuracy.Auto-MatRegressor supports dynamic expansion of meta-data with the increase of the number of materials datasets and other required algorithms and can be applied to any ML materials discovery and designtask.