A general and transferable deep learning framework for predicting phase formation in materials

Shuo Feng,Huadong Fu,Yuan Wu,Zhaoping Lu,Hongbiao Dong,Huiyu Zhou

doi:10.1038/s41524-020-00488-z

Shuo Feng, Huadong Fu + Show 4 more

Open Access

https://doi.org/10.1038/s41524-020-00488-z

Copy DOI

Abstract

Machine learning has been widely exploited in developing new materials. However, challenges still exist: small dataset is common for most tasks; new datasets, special descriptors and specific models need to be built from scratch when facing a new task; knowledge cannot be readily transferred between independent models. In this paper we propose a general and transferable deep learning (GTDL) framework for predicting phase formation in materials. The proposed GTDL framework maps raw data to pseudo-images with some special 2-D structure, e.g., periodic table, automatically extracts features and gains knowledge through convolutional neural network, and then transfers knowledge by sharing features extractors between models. Application of the GTDL framework in case studies on glass-forming ability and high-entropy alloys show that the GTDL framework for glass-forming ability outperformed previous models and can correctly predicted the newly reported amorphous alloy systems; for high-entropy alloys the GTDL framework can discriminate five types phases (BCC, FCC, HCP, amorphous, mixture) with accuracy and recall above 94% in fivefold cross-validation. In addition, periodic table knowledge embedded in data representations and knowledge shared between models is beneficial for tasks with small dataset. This method can be easily applied to new materials development with small dataset by reusing well-trained models for related materials.

Highlights

Machine learning is a powerful tool which has become an important complement to experiment, theory, and modeling[1,2,3,4,5,6]
Ward et al.[7] first used 145 generalpurpose Magpie descriptors in predicting ternary amorphous ribbon alloys (AMRs). Later they used 210 descriptors in optimizing Zr-based bulk metallic glass (BMG)[33]
We propose a general and transferable deep learning models using manual feature engineering (see the full learning (GTDL) framework to predict phase formation in materials list of the features in Supplementary Table 1) to validate the with small dataset and unclear transformation mechanism

Summary

INTRODUCTION

Machine learning is a powerful tool which has become an important complement to experiment, theory, and modeling[1,2,3,4,5,6]. We propose a general and transferable deep learning models using manual feature engineering (see the full learning (GTDL) framework to predict phase formation in materials list of the features in Supplementary Table 1) to validate the with small dataset and unclear transformation mechanism. A clear (containing 10000+ pieces of data) of GFA and a small dataset advantage of deep learning framework over conventional (containing only 355 pieces of data) of HEAs demonstrate: GTDL machine learning is it can automatically extract features and framework outperforms existing models based on manual transfer knowledge. The proposed GDTL framework can be used in new materials development with small datasets by exploiting trained deep structures to reduce the risk of overfitting limited data in our tasks.

RESULTS

METHODS

CODE AVAILABILITY