mlsanityData to 7 Oct 2026

Bayesian OOD detection with aleatoric uncertainty and outlier exposure

arXiv 2102.12959 ↗First posted 24 Feb 2021stat.ML cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

Typical Bayesian approaches to OOD detection use epistemic uncertainty. Surprisingly from the Bayesian perspective, there are a number of methods that successfully use aleatoric uncertainty to detect OOD points (e.g. Hendryks et al. 2018). In addition, it is difficult to use outlier exposure to improve a Bayesian OOD detection model, as it is not clear whether it is possible or desirable to increase posterior (epistemic) uncertainty at outlier points. We show that a generative model of data curation provides a principled account of aleatoric uncertainty for OOD detection. In particular, aleatoric uncertainty signals a specific type of OOD point: one without a well-defined class-label, and our model of data curation gives a likelihood for these points, giving us a mechanism for conditioning on outlier points and thus performing principled Bayesian outlier exposure. Our principled Bayesian approach, combining aleatoric and epistemic uncertainty with outlier exposure performs better than methods using aleatoric or epistemic alone.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 30 of its 49 references that are in the map. First posted on arXiv 24 Feb 2021.

Builds on

30 of 49 references.

References · 30

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  3. 2016Wide Residual Networkscited by 3,057
  4. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  5. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,335
  6. 2013Describing Textures in the Wildcited by 2,110
  7. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,274
  8. 2018Deep Anomaly Detection with Outlier Exposurecited by 939
  9. 2016Pixel Recurrent Neural Networkscited by 814
  10. 2018Predictive Uncertainty Estimation via Prior Networkscited by 555
  11. 2019A Simple Baseline for Bayesian Uncertainty in Deep Learningcited by 555
  12. 2019Likelihood Ratios for Out-of-Distribution Detectioncited by 451
18 more
  1. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  2. 2017Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samplescited by 389
  3. 2020BatchEnsemble: An Alternative Approach to Efficient Ensemble and Lifelong Learningcited by 368
  4. 2020Consistent Estimators for Learning to Defer to an Expertcited by 262
  5. 2018Learning Confidence for Out-of-Distribution Detection in Neural Networkscited by 245
  6. 2019Cyclical Stochastic Gradient MCMC for Bayesian Deep Learningcited by 203
  7. 2018Reducing Network Agnostophobiacited by 165
  8. 2019Ensemble Distribution Distillationcited by 158
  9. 2017Multiplicative Normalizing Flows for Variational Bayesian Neural Networkscited by 157
  10. 2021Bayesian Neural Network Priors Revisitedcited by 121
  11. 2018Generative Probabilistic Novelty Detection with Adversarial Autoencoderscited by 99
  12. 2020Density of States Estimation for Out-of-Distribution Detectioncited by 74
  13. 2020A statistical theory of cold posteriors in deep neural networkscited by 53
  14. 2019Out-of-distribution Detection in Classifiers via Generationcited by 49
  15. 2018Open Category Detection with PAC Guaranteescited by 35
  16. 2017Implicit Weight Uncertainty in Neural Networkscited by 30
  17. 2018A Less Biased Evaluation of Out-of-distribution Sample Detectorscited by 14
  18. 2017Safer Classification by Synthesiscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 3; 2023: 2; 2024: 3; 2025: 5; 2026: 1.

16 papers.

Sort:

By area:

  1. 2024Rethinking Aleatoric and Epistemic Uncertaintycited by 40
  2. 2021Data augmentation in Bayesian neural networks and the cold posterior effectcited by 36
  3. 2022BayesCap: Bayesian Identity Cap for Calibrated Uncertainty in Frozen Neural Networkscited by 23
  4. 2021Being a Bit Frequentist Improves Bayesian Neural Networkscited by 15
  5. 2025Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understandingcited by 13
  6. 2022Towards out of distribution generalization for problems in mechanicscited by 11
  7. 2023ProbVLM: Probabilistic Adapter for Frozen Vision-Language Modelscited by 7
  8. 2023Out-of-Distribution Detection for Monocular Depth Estimationcited by 7
  9. 2022Robustness to corruption in pre-trained Bayesian neural networkscited by 5
  10. 2025EigenScore: OOD Detection using Covariance in Diffusion Modelscited by 4
  11. 2024Fully Heteroscedastic Count Regression with Deep Double Poisson Networkscited by 3
  12. 2024PViT: Prior-augmented Vision Transformer for Out-of-distribution Detectioncited by 3
4 more
  1. 2025Logit Disagreement: OoD Detection with Bayesian Neural Networkscited by 0
  2. 2025Out-of-Distribution Detection from Small Training Sets using Bayesian Neural Network Classifierscited by 0
  3. 2025Dynamic Correction of Erroneous State Estimates via Diffusion Bayesian Explorationcited by 0
  4. 2026Uncertainty quantification for trustworthy deep learning: Methods and measurescited by 0