# Quantifying Out-of-Training Uncertainty of Neural-Network based Turbulence Closures

arXiv: https://arxiv.org/abs/2508.16891 · HTML: https://mlsanity.com/paper/2508.16891/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 53 references that are in the map.
- First posted on arXiv 23 Aug 2025.

## Builds on (most cited first, 16 of 16)

- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [A Survey of Uncertainty in Deep Neural Networks](https://mlsanity.com/paper/2107.03342/) (2021)
- [GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration](https://mlsanity.com/paper/1809.11165/) (2018)
- [Deep Ensembles: A Loss Landscape Perspective](https://mlsanity.com/paper/1912.02757/) (2019)
- [Accurate Uncertainties for Deep Learning Using Calibrated Regression](https://mlsanity.com/paper/1807.00263/) (2018)
- [What Are Bayesian Neural Network Posteriors Really Like?](https://mlsanity.com/paper/2104.14421/) (2021)
- [Concrete Dropout](https://mlsanity.com/paper/1705.07832/) (2017)
- [Enhancing Computational Fluid Dynamics with Machine Learning](https://mlsanity.com/paper/2110.02085/) (2021)
- [Modeling the Dynamics of PDE Systems with Physics-Constrained Deep Auto-Regressive Networks](https://mlsanity.com/paper/1906.05747/) (2019)
- [Repulsive Deep Ensembles are Bayesian](https://mlsanity.com/paper/2106.11642/) (2021)
- [All You Need is a Good Functional Prior for Bayesian Deep Learning](https://mlsanity.com/paper/2011.12829/) (2020)
- [Deep Ensembles from a Bayesian Perspective](https://mlsanity.com/paper/2105.13283/) (2021)
- [Quantifying model form uncertainty in Reynolds-averaged turbulence models with Bayesian deep neural networks](https://mlsanity.com/paper/1807.02901/) (2018)
- [Empowering Bayesian Neural Networks with Functional Priors through Anchored Ensembling for Mechanics Surrogate Modeling Applications](https://mlsanity.com/paper/2409.05234/) (2024)
- [Uncertainty Quantification for Data-Driven Machine Learning Models in Nuclear Engineering Applications: Where We Are and What Do We Need?](https://mlsanity.com/paper/2503.17385/) (2025)

## Abstract (arXiv)

Neural-Network (NN) based turbulence closures have been developed for being used as pre-trained surrogates for traditional turbulence closures, with the aim to increase computational efficiency and prediction accuracy of CFD simulations. The bottleneck to the widespread adaptation of these ML-based closures is the relative lack of uncertainty quantification (UQ) for these models. Especially, quantifying uncertainties associated with out-of-training inputs, that is when the ML-based turbulence closures are queried on inputs outside their training data regime. In the current paper, a published algebraic turbulence closure1 has been utilized to compare the quality of epistemic UQ between three NN-based methods and Gaussian Process (GP). The three NN-based methods explored are Deep Ensembles (DE), Monte-Carlo Dropout (MCD), and Stochastic Variational Inference (SVI). In the in-training results, we find the exact GP performs the best in accuracy with a Root Mean Squared Error (RMSE) of 2.14 · 10-5 followed by the DE with an RMSE of 4.59 · 10-4. Next, the paper discusses the performance of the four methods for quantifying out-of-training uncertainties. For performance, the Exact GP yet again is the best in performance, but has similar performance to the DE in the out-of-training regions. In UQ accuracy for the out-of-training case, SVI and DE hold the best miscalibration error for one of the cases. However, the DE performs the best in Negative Log-Likelihood for both out-of-training cases. We observe that for the current problem, in terms of accuracy GP > DE > SV I > MCD. The DE results are relatively robust and provide intuitive UQ estimates, despite performing naive ensembling. In terms of computational cost, the GP is significantly higher than the NN-based methods with a O(n3) computational complexity for each training step