The Rise of Neural Networks for Materials and Chemical Dynamics

Machine learning (ML) is quickly becoming a premier tool for modeling chemical processes and materials. ML-based force fields, trained on large data sets of high-quality electron structure calculations, are particularly attractive due their unique combination of computational efficiency and physical accuracy. This Perspective summarizes some recent advances in the development of neural network-based interatomic potentials. Designing high-quality training data sets is crucial to overall model accuracy. One strategy is active learning, in which new data are automatically collected for atomic configurations that produce large ML uncertainties. Another strategy is to use the highest levels of quantum theory possible. Transfer learning allows training to a data set of mixed fidelity. A model initially trained to a large data set of density functional theory calculations can be significantly improved by retraining to a relatively small data set of expensive coupled cluster theory calculations. These advances are exemplified by applications to molecules and materials.

Automated summary

Machine learning is rapidly becoming a key tool for modeling chemical processes and materials, especially with the development of ML-based force fields for interatomic potentials. Designing high-quality training data sets is crucial for model accuracy, with strategies like active learning and using high levels of quantum theory to enhance training. Transfer learning allows for training on mixed fidelity data sets, showing significant improvements in model accuracy.