Prediction-accuracy improvement of neural network to ferromagnetic multilayers by Gaussian data augmentation and ensemble learning

• Published in: Computational materials science Vol. 219; p. 112032
• Main Authors: Nawa, Kenji, Hagiwara, Katsuyuki, Nakamura, Kohji
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 25-02-2023
• Subjects: Ensemble learning; First-principles calculations; Gaussian data augmentation method; Magnetic multilayers; Neural network; First-principles calculations; Magnetic multilayers; Neural network; Ensemble learning; Gaussian data augmentation method
• ISSN: 0927-0256; 1879-0801
• DOI: 10.1016/j.commatsci.2023.112032

In materials informatics using machine learning and density functional theory (DFT) calculations, it is often hard to obtain enough database due to extremely large costs of DFT. Therefore, it is required a machine learning technique that learns a complex target relationship from a limited dataset. In the present work, to overcome this issue, we built a neural network by implementing two techniques: Gaussian data augmentation (GDA) method, which injects Gaussian noises into the training dataset, and ensemble learning, which employs multiple models to train and make prediction by averaging their outputs. With typical examples of magnetic moment and formation energy as a function of atomic-layer configuration in CoFe multilayers, the prediction accuracy can be greatly improved, e.g., by using a training dataset consisting of 10∼30% of all data. We found that the use of GDA substantially increases the prediction accuracy for unknown test dataset where the improvement is attributed to a smoothing effect of a fitting curve of NN, and a combination with the ensemble learning brings further improvement with reducing the variance error originating from the selection of training sampling dataset in addition to a similar smoothing effect. The present approach, thus, can be generalized widely to materials informatics for which database is limited.
[Display omitted]
•Machine-learning neural network (NN) is applied to CoFe magnetic multilayers.•Prediction accuracy can be greatly improved by using Gaussian data augmentation.•The improvement is attributed to a smoothing effect of NN fitting curve.•Ensemble learning for the NN also yields a similar smoothing effect.•The present NN can be generalized widely to materials informatics in the future.