Fast and Accurate Uncertainty Estimation in Chemical Machine Learning.

By Félix Musil, M. J. Willatt, Mikhail A. Langovoy, Michele Ceriotti. 2018

148 citations recorded by the source corpus.

We present a scheme to obtain an inexpensive and reliable estimate of the uncertainty associated with the predictions of a machine-learning model of atomic and molecular properties. The scheme is based on resampling, with multiple models being generated based on subsampling of the same training data. The accuracy of the uncertainty prediction can be benchmarked by maximum likelihood estimation, which can also be used to correct for correlations between resampled models and to improve the performance of the uncertainty estimation by a cross-validation procedure. In the case of sparse Gaussian Process Regression models, this resampled estimator can be evaluated at negligible cost. We demonstrate the reliability of these estimates for the prediction of molecular and materials energetics and for the estimation of nuclear chemical shieldings in molecular crystals. Extension to estimate the uncertainty in energy differences, forces, or other correlated predictions is straightforward. This method can be easily applied to other machine-learning schemes and will be beneficial to make data-driven predictions more reliable and to facilitate training-set optimization and active-learning strategies.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Rohit Batra
2. Willie Neiswanger
3. Eric P. Xing
4. E. Weinan
5. Le Song
6. Han Wang
7. Linfeng Zhang
8. Roberto Car

In Materials Chemistry

1. Linfeng Zhang
2. Han Wang
3. E. Weinan
4. Rohit Batra
5. Roberto Car
6. Rampi Ramprasad
7. Eric P. Xing
8. Le Song

In Molecular Biology

1. Han Wang
2. Linfeng Zhang
3. Rohit Batra
4. Eric P. Xing
5. E. Weinan
6. Le Song
7. Willie Neiswanger
8. Rampi Ramprasad

Citations in the corpus

• Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation (2018, background): “by recent work by other groups who used closely related indicators in applications to a variety of different systems [30, 33].”
• Methods for comparing uncertainty quantifications for material property predictions (2019, methodology): “These methods are all underpinned by models with UQ, which have garnered increasing attention [13, 14].”
• Emerging materials intelligence ecosystems propelled by machine learning (2020, background): “Bayesian-based uncertainty estimates 154 or estimates based on deviations of energy and force predictions from ensemble methods 155 have shown promise.”