Key feature space for predicting the glass-forming ability of amorphous alloys revealed by gradient boosted decision trees model

The glass forming ability (GFA) is a problem of great concern in the research of amorphous materials. It is of great significance to understand the physical mechanism of GFA and to seek conditions and methods to improve it. In this study, we collected 820 experimental data from existing literature, and used gradient boosted decision trees (GBDT) model to predict the GFA. The GBDT model optimized by 10-fold crossvalidation and grid search technology shows excellent predictive results. The determination coefficient (R2) and root mean square error (RMSE) are 0.652 and 2.85, respectively. Compared with the previously reported 27 criteria and ML models, GBDT model has the highest prediction ability. The result exhibit that the predictive performance of GBDT can be significantly improved by considering the atomic size difference, total electronegativity, mixing entropy and average atomic volume. (c) 2022 Elsevier B.V. All rights reserved.

Automated summary

The glass forming ability (GFA) of amorphous materials is of great concern. In this study, a gradient boosted decision trees (GBDT) model was used to predict GFA based on 820 experimental data, achieving excellent predictive results. The GBDT model shows the highest prediction ability compared to previous criteria and ML models.