Materials Property Prediction with Uncertainty Quantification: A Benchmark Study
This work addresses the need for reliable uncertainty estimation in materials science, which is crucial for robust predictions and active learning, but it is incremental as it benchmarks existing methods.
The study tackled uncertainty quantification in materials property prediction by evaluating various methods on four crystal materials datasets, finding that popular ensemble methods are not the best for this task.
Uncertainty quantification (UQ) has increasing importance in building robust high-performance and generalizable materials property prediction models. It can also be used in active learning to train better models by focusing on getting new training data from uncertain regions. There are several categories of UQ methods each considering different types of uncertainty sources. Here we conduct a comprehensive evaluation on the UQ methods for graph neural network based materials property prediction and evaluate how they truly reflect the uncertainty that we want in error bound estimation or active learning. Our experimental results over four crystal materials datasets (including formation energy, adsorption energy, total energy, and band gap properties) show that the popular ensemble methods for uncertainty estimation is NOT the best choice for UQ in materials property prediction. For the convenience of the community, all the source code and data sets can be accessed freely at \url{https://github.com/usccolumbia/materialsUQ}.