Calibrated Uncertainty for Molecular Property Prediction using Ensembles of Message Passing Neural Networks

TL;DR

This paper introduces a method for molecular property prediction using message passing neural networks that provides well-calibrated uncertainty estimates by considering both aleatoric and epistemic uncertainties and recalibrating on unseen data.

Contribution

It extends message passing neural networks to produce calibrated probabilistic predictions by integrating uncertainty types and recalibrating on new data, improving reliability.

Findings

Accurate molecular property predictions with calibrated uncertainty on QM9 and PC9 datasets.

Unified framework for aleatoric and epistemic uncertainty estimation.

Recalibration improves uncertainty estimates on unseen data.

Abstract

Data-driven methods based on machine learning have the potential to accelerate computational analysis of atomic structures. In this context, reliable uncertainty estimates are important for assessing confidence in predictions and enabling decision making. However, machine learning models can produce badly calibrated uncertainty estimates and it is therefore crucial to detect and handle uncertainty carefully. In this work we extend a message passing neural network designed specifically for predicting properties of molecules and materials with a calibrated probabilistic predictive distribution. The method presented in this paper differs from previous work by considering both aleatoric and epistemic uncertainty in a unified framework, and by recalibrating the predictive distribution on unseen data. Through computer experiments, we show that our approach results in accurate models for predicting molecular formation energies with well calibrated uncertainty in and out of the training data distribution on two public molecular benchmark datasets, QM9 and PC9. The proposed method provides a general framework for training and evaluating neural network ensemble models that are able to produce accurate predictions of properties of molecules with well calibrated uncertainty estimates.