mlsanityData to 7 Oct 2026

Uncertainty Estimation for Molecules: Desiderata and Methods

arXiv 2306.14916 ↗First posted 20 Jun 2023physics.chem-ph cs.LG physics.comp-phcited by 11

Not on the map. Why?

Abstract (arXiv)

Graph Neural Networks (GNNs) are promising surrogates for quantum mechanical calculations as they establish unprecedented low errors on collections of molecular dynamics (MD) trajectories. Thanks to their fast inference times they promise to accelerate computational chemistry applications. Unfortunately, despite low in-distribution (ID) errors, such GNNs might be horribly wrong for out-of-distribution (OOD) samples. Uncertainty estimation (UE) may aid in such situations by communicating the model's certainty about its prediction. Here, we take a closer look at the problem and identify six key desiderata for UE in molecular force fields, three 'physics-informed' and three 'application-focused' ones. To overview the field, we survey existing methods from the field of UE and analyze how they fit to the set desiderata. By our analysis, we conclude that none of the previous works satisfies all criteria. To fill this gap, we propose Localized Neural Kernel (LNK) a Gaussian Process (GP)-based extension to existing GNNs satisfying the desiderata. In our extensive experimental evaluation, we test four different UE with three different backbones and two datasets. In out-of-equilibrium detection, we find LNK yielding up to 2.5 and 2.1 times lower errors in terms of AUC-ROC score than dropout or evidential regression-based methods while maintaining high predictive performance.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 51 of its 86 references that are in the map. First posted on arXiv 20 Jun 2023.

Builds on

51 of 86 references.

References · 51

  1. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  2. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  3. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,274
  4. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
  5. 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 701
  6. 2021A Survey of Uncertainty in Deep Neural Networkscited by 675
  7. 2020Directional Message Passing for Molecular Graphscited by 556
  8. 2019A Simple Baseline for Bayesian Uncertainty in Deep Learningcited by 555
  9. 2015Weight Uncertainty in Neural Networkscited by 554
  10. 2020A Survey of Deep Active Learningcited by 547
  11. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 438
  12. 2018Accurate Uncertainties for Deep Learning Using Calibrated Regressioncited by 417
38 more
  1. 2020Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awarenesscited by 389
  2. 2013Gaussian Processes for Big Datacited by 377
  3. 2020BatchEnsemble: An Alternative Approach to Efficient Ensemble and Lifelong Learningcited by 368
  4. 2019Deep Evidential Regressioncited by 354
  5. 2021GemNet: Universal Directional Graph Neural Networks for Moleculescited by 342
  6. 2015Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networkscited by 295
  7. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 254
  8. 2014Scalable Variational Gaussian Process Classificationcited by 226
  9. 2021Spherical Message Passing for 3D Graph Networkscited by 166
  10. 2020Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Countscited by 163
  11. 2020Hyperparameter Ensembles for Robustness and Uncertainty Quantificationcited by 149
  12. 2020Confidence-Aware Learning for Deep Neural Networkscited by 113
  13. 2021SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effectscited by 103
  14. 2022The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentialscited by 100
  15. 2022Out-Of-Distribution Generalization on Graphs: A Surveycited by 97
  16. 2022Forces are not Enough: Benchmark and Critical Evaluation for Machine Learning Force Fields with Molecular Simulationscited by 92
  17. 2019Distribution Calibration for Regressioncited by 88
  18. 2021On Feature Collapse and Deep Kernel Learning for Single Forward Pass Uncertaintycited by 87
  19. 2021The Promises and Pitfalls of Deep Kernel Learningcited by 81
  20. 2021Graph Posterior Network: Bayesian Predictive Uncertainty for Node Classificationcited by 79
  21. 2018Lightweight Probabilistic Deep Networkscited by 77
  22. 2019Sampling-free Epistemic Uncertainty Estimation Using Approximated Variance Propagationcited by 74
  23. 2020Intra Order-preserving Functions for Calibration of Multi-Class Neural Networkscited by 65
  24. 2020OrbNet: Deep Learning for Quantum Chemistry Using Symmetry-Adapted Atomic-Orbital Featurescited by 63
  25. 2021Smooth Normalizing Flowscited by 52
  26. 2020Individual Calibration with Randomized Forecastingcited by 50
  27. 2020Uncertainty-Aware Deep Classifiers using Generative Modelscited by 41
  28. 2020Towards Maximizing the Representation Gap between In-Domain & Out-of-Distribution Examplescited by 40
  29. 2020Evaluating Robustness of Predictive Uncertainty Estimation: Are Dirichlet-based Models Reliable?cited by 39
  30. 2019Parting with Illusions about Deep Active Learningcited by 38
  31. 2021Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave Functionscited by 36
  32. 2016Natural-Parameter Networks: A Class of Probabilistic Neural Networkscited by 33
  33. 2019Uncertainty on Asynchronous Time Event Predictioncited by 27
  34. 2023Generalizing Neural Wave Functionscited by 26
  35. 2022Sampling-free Inference for Ab-Initio Potential Energy Surface Networkscited by 19
  36. 2021Natural Posterior Network: Deep Bayesian Uncertainty for Exponential Family Distributionscited by 13
  37. 2021Deep Gaussian Processes: A Surveycited by 13
  38. 2020Graph Mixture Density Networkscited by 9

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 4; 2026: 2.

11 papers.

  1. 2024Uncertainty in Graph Neural Networks: A Surveycited by 20
  2. 2025Uncertainty Quantification for Machine Learning in Healthcare: A Surveycited by 11
  3. 2025chemtrain-deploy: A parallel and scalable framework for machine learning potentials in million-atom MD simulationscited by 6
  4. 2024LTAU-FF: Loss Trajectory Analysis for Uncertainty in Atomistic Force Fieldscited by 3
  5. 2024Uncertainty Quantification on Graph Learning: A Surveycited by 3
  6. 2024On Representing Electronic Wave Functions with Sign Equivariant Neural Networkscited by 2
  7. 2024AUGUR, A flexible and efficient optimization algorithm for identification of optimal adsorption sitescited by 2
  8. 2025On Equivariant Model Selection through the Lens of Uncertaintycited by 1
  9. 2025Two Birds with One Stone: Enhancing Uncertainty Quantification and Interpretability with Graph Functional Neural Processcited by 0
  10. 2026Equivariant Evidential Deep Learning for Interatomic Potentialscited by 0
  11. 2026VINCENT: Validated Interaction Network for Cross-drug Explanation of Therapeuticscited by 0