Improved Uncertainty Estimation of Graph Neural Network Potentials Using Engineered Latent Space Distances

TL;DR

This paper enhances uncertainty estimation in graph neural network potentials for molecular energy prediction by developing a latent space distance method, improving calibration and applicability in material discovery tasks.

Contribution

It introduces a novel latent space distance approach for uncertainty quantification in GNNs, specifically tailored for relaxed energy calculations, with improved calibration and efficiency.

Findings

Latent distance methods are most well-calibrated for relaxed energy predictions.

The proposed method is more economical and effective than existing uncertainty methods.

Results align with expectations on clustering and outside dataset examples.

Abstract

Graph neural networks (GNNs) have been shown to be astonishingly capable models for molecular property prediction, particularly as surrogates for expensive density functional theory calculations of relaxed energy for novel material discovery. However, one limitation of GNNs in this context is the lack of useful uncertainty prediction methods, as this is critical to the material discovery pipeline. In this work, we show that uncertainty quantification for relaxed energy calculations is more complex than uncertainty quantification for other kinds of molecular property prediction, due to the effect that structure optimizations have on the error distribution. We propose that distribution-free techniques are more useful tools for assessing calibration, recalibrating, and developing uncertainty prediction methods for GNNs performing relaxed energy calculations. We also develop a relaxed energy task for evaluating uncertainty methods for equivariant GNNs, based on distribution-free recalibration and using the Open Catalyst Project dataset. We benchmark a set of popular uncertainty prediction methods on this task, and show that latent distance methods, with our novel improvements, are the most well-calibrated and economical approach for relaxed energy calculations. Finally, we demonstrate that our latent space distance method produces results which align with our expectations on a clustering example, and on specific equation of state and adsorbate coverage examples from outside the training dataset.