# Are Bayesian neural networks intrinsically good at out-of-distribution detection?

arXiv: https://arxiv.org/abs/2107.12248 · HTML: https://mlsanity.com/paper/2107.12248/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 29 references that are in the map.
- First posted on arXiv 26 Jul 2021.

## Builds on (most cited first, 11 of 11)

- [Normalizing Flows for Probabilistic Modeling and Inference](https://mlsanity.com/paper/1912.02762/) (2019)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Deep Neural Networks as Gaussian Processes](https://mlsanity.com/paper/1711.00165/) (2017)
- [A Simple Baseline for Bayesian Uncertainty in Deep Learning](https://mlsanity.com/paper/1902.02476/) (2019)
- [Bayesian Deep Learning and a Probabilistic Perspective of Generalization](https://mlsanity.com/paper/2002.08791/) (2020)
- [Do Deep Generative Models Know What They Don't Know?](https://mlsanity.com/paper/1810.09136/) (2018)
- [A Kernelized Stein Discrepancy for Goodness-of-fit Tests and Model Evaluation](https://mlsanity.com/paper/1602.03253/) (2016)
- [Multiplicative Normalizing Flows for Variational Bayesian Neural Networks](https://mlsanity.com/paper/1703.01961/) (2017)
- [Bayesian Hypernetworks](https://mlsanity.com/paper/1710.04759/) (2017)
- [Posterior Meta-Replay for Continual Learning](https://mlsanity.com/paper/2103.01133/) (2021)
- [Implicit Weight Uncertainty in Neural Networks](https://mlsanity.com/paper/1711.01297/) (2017)

## Built on by (most cited first, 10 of 10)

- [Towards out of distribution generalization for problems in mechanics](https://mlsanity.com/paper/2206.14917/) (2022)
- [Are We Ready for Out-of-Distribution Detection in Digital Pathology?](https://mlsanity.com/paper/2407.13708/) (2024)
- [Calibrating Bayesian Learning via Regularization, Confidence Minimization, and Selective Inference](https://mlsanity.com/paper/2404.11350/) (2024)
- [PViT: Prior-augmented Vision Transformer for Out-of-distribution Detection](https://mlsanity.com/paper/2410.20631/) (2024)
- [Uncertainty estimation under model misspecification in neural network regression](https://mlsanity.com/paper/2111.11763/) (2021)
- [Bayesian deep learning for cosmic volumes with modified gravity](https://mlsanity.com/paper/2309.00612/) (2023)
- [Distributional Gaussian Processes Layers for Out-of-Distribution Detection](https://mlsanity.com/paper/2206.13346/) (2022)
- [An out-of-distribution discriminator based on Bayesian neural network epistemic uncertainty](https://mlsanity.com/paper/2210.10780/) (2022)
- [Generalization is not a universal guarantee: Estimating similarity to training data with an ensemble out-of-distribution metric](https://mlsanity.com/paper/2502.16329/) (2025)
- [Out-of-Distribution Detection from Small Training Sets using Bayesian Neural Network Classifiers](https://mlsanity.com/paper/2510.06025/) (2025)

## Abstract (arXiv)

The need to avoid confident predictions on unfamiliar data has sparked interest in out-of-distribution (OOD) detection. It is widely assumed that Bayesian neural networks (BNN) are well suited for this task, as the endowed epistemic uncertainty should lead to disagreement in predictions on outliers. In this paper, we question this assumption and provide empirical evidence that proper Bayesian inference with common neural network architectures does not necessarily lead to good OOD detection. To circumvent the use of approximate inference, we start by studying the infinite-width case, where Bayesian inference can be exact considering the corresponding Gaussian process. Strikingly, the kernels induced under common architectural choices lead to uncertainties that do not reflect the underlying data generating process and are therefore unsuited for OOD detection. Finally, we study finite-width networks using HMC, and observe OOD behavior that is consistent with the infinite-width case. Overall, our study discloses fundamental problems when naively using BNNs for OOD detection and opens interesting avenues for future research.