Interpretable Ensemble Learning for Materials Property Prediction with Classical Interatomic Potentials: Carbon as an Example
This work addresses the need for efficient and interpretable property prediction in materials science, though it is incremental as it applies ensemble learning to a specific domain with small datasets.
The paper tackled the problem of predicting materials properties like formation energy and elastic constants for carbon allotropes using machine learning, by proposing an ensemble learning approach with regression trees that avoids descriptors and uses inputs from classical interatomic potentials, resulting in more accurate predictions than the classical potentials alone.
Machine learning (ML) is widely used to explore crystal materials and predict their properties. However, the training is time-consuming for deep-learning models, and the regression process is a black box that is hard to interpret. Also, the preprocess to transfer a crystal structure into the input of ML, called descriptor, needs to be designed carefully. To efficiently predict important properties of materials, we propose an approach based on ensemble learning consisting of regression trees to predict formation energy and elastic constants based on small-size datasets of carbon allotropes as an example. Without using any descriptor, the inputs are the properties calculated by molecular dynamics with 9 different classical interatomic potentials. Overall, the results from ensemble learning are more accurate than those from classical interatomic potentials, and ensemble learning can capture the relatively accurate properties from the 9 classical potentials as criteria for predicting the final properties.