mlsanityData to 14 Oct 2026

Predictive Uncertainty for Neural CAE Surrogates

arXiv 2609.25430 ↗First posted 21 Sep 2026cs.LG cs.AI math.PRcited by 0

Not on the map. Why?

Abstract (arXiv)

Neural surrogates can substantially accelerate computer-aided engineering (CAE) workflows, but their use in design requires uncertainty estimates that remain meaningful across varying geometries, spatial prediction fields, and engineering quantities of interest. We investigate how established uncertainty quantification (UQ) approaches behave when adapted to geometry-conditioned neural surrogates. We compare one closed-form and two sampling-based approaches-a Gaussian process (GP)-based method, concrete Monte Carlo (MC) dropout, and deep ensembles-and evaluate them on three large, industry-relevant CAE datasets for external aerodynamics and crash dynamics. We examine whether predicted uncertainties have credible magnitudes, identify locations with larger prediction errors, respond to unfamiliar inputs, and remain informative for derived engineering quantities. On the DrivAerStar dataset, where all three methods are compared, each generally assigns higher uncertainty to locations with larger prediction errors, and validation-based rescaling brings interval coverage close to nominal on a disjoint in-distribution test set. Results on AirFRANS and automotive crash also show useful error ranking and interval estimates, but the relative performance of the methods changes with the dataset and evaluation criterion. UQ methods and evaluation metrics should therefore be selected based on the intended downstream CAE decision.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 30 of its 41 references that are in the map. First posted on arXiv 21 Sep 2026.

Builds on

30 of 41 references.

References · 30

  1. 2017On Calibration of Modern Neural Networkscited by 4,960
  2. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  3. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  4. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,908
  5. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  6. 2019DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operatorscited by 2,018
  7. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,287
  8. 2017Selective Classification for Deep Neural Networkscited by 852
  9. 2017Deep Bayesian Active Learning with Image Datacited by 819
  10. 2015Deep Kernel Learningcited by 445
  11. 2018Accurate Uncertainties for Deep Learning Using Calibrated Regressioncited by 422
  12. 2020Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awarenesscited by 389
18 more
  1. 2024Transolver: A Fast Transformer Solver for PDEs on General Geometriescited by 326
  2. 2014Scalable Variational Gaussian Process Classificationcited by 228
  3. 2017Concrete Dropoutcited by 227
  4. 2018Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantificationcited by 199
  5. 2022Uncertainty Quantification in Scientific Machine Learning: Methods, Metrics, and Comparisonscited by 170
  6. 2018Uncertainty Estimates and Multi-Hypotheses Networks for Optical Flowcited by 135
  7. 2025Transolver++: An Accurate Neural Solver for PDEs on Million-Scale Geometriescited by 97
  8. 2021On Feature Collapse and Deep Kernel Learning for Single Forward Pass Uncertaintycited by 87
  9. 2022AirfRANS: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier-Stokes Solutionscited by 82
  10. 2025DoMINO: A Decomposable Multi-scale Iterative Neural Operator for Modeling Large Scale Engineering Simulationscited by 27
  11. 2025GeoTransolver: Learning Physics on Irregular Domains Using Multi-scale Geometry Aware Physics Attention Transformercited by 23
  12. 2018Quantifying model form uncertainty in Reynolds-averaged turbulence models with Bayesian deep neural networkscited by 20
  13. 2024Linearization Turns Neural Operators into Function-Valued Gaussian Processescited by 12
  14. 2025Automotive Crash Dynamics Modeling Accelerated with Machine Learningcited by 10
  15. 2025Calibrated Physics-Informed Uncertainty Quantificationcited by 8
  16. 2025DrivAerStar: An Industrial-Grade CFD Dataset for Vehicle Aerodynamic Optimizationcited by 7
  17. 2025A Benchmarking Framework for AI models in Automotive Aerodynamicscited by 4
  18. 2026High-Fidelity Industrial Crash Dynamics Prediction via Geometry-Aware Operator Learning with Memory-Efficient Low-Rank Attentioncited by 4

Built on by

Not cited yet.