mlsanityData to 14 Oct 2026

Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods?

arXiv 2402.09056 ↗First posted 14 Feb 2024cs.AI cs.LGcited by 35

Not on the map. Why?

Abstract (arXiv)

Trustworthy ML systems should not only return accurate predictions, but also a reliable representation of their uncertainty. Bayesian methods are commonly used to quantify both aleatoric and epistemic uncertainty, but alternative approaches, such as evidential deep learning methods, have become popular in recent years. The latter group of methods in essence extends empirical risk minimization (ERM) for predicting second-order probability distributions over outcomes, from which measures of epistemic (and aleatoric) uncertainty can be extracted. This paper presents novel theoretical insights of evidential deep learning, highlighting the difficulties in optimizing second-order loss functions and interpreting the resulting epistemic uncertainty measures. With a systematic setup that covers a wide range of approaches for classification, regression and counts, it provides novel insights into issues of identifiability and convergence in second-order loss minimization, and the relative (rather than absolute) nature of epistemic uncertainty measures.

Key facts. Cited by 35 papers in the map (data through 14 Oct 2026). Builds on 29 of its 40 references that are in the map. First posted on arXiv 14 Feb 2024.

Builds on

29 of 40 references.

References · 29

  1. 2021Convex optimizationcited by 2,515
  2. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  3. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 913
  4. 2018Evidential Deep Learning to Quantify Classification Uncertaintycited by 732
  5. 2018Predictive Uncertainty Estimation via Prior Networkscited by 557
  6. 2019Your Classifier is Secretly an Energy Based Model and You Should Treat it Like Onecited by 405
  7. 2019Deep Evidential Regressioncited by 355
  8. 2022OpenOOD: Benchmarking Generalized Out-of-Distribution Detectioncited by 268
  9. 2013A General Framework for Updating Belief Distributionscited by 175
  10. 2020Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Countscited by 164
  11. 2022Quantifying Aleatoric and Epistemic Uncertainty in Machine Learning: Are Conditional Entropy and Mutual Information Appropriate Measures?cited by 125
  12. 2019Reverse KL-Divergence Training of Prior Networks: Improved Uncertainty and Adversarial Robustnesscited by 104
17 more
  1. 2021Evidential Deep Learning for Open Set Action Recognitioncited by 96
  2. 2021Prior and Posterior Networks: A Survey on Evidential Deep Learning Methods For Uncertainty Estimationcited by 71
  3. 2022Pitfalls of Epistemic Uncertainty Quantification through Loss Minimisationcited by 66
  4. 2022The Unreasonable Effectiveness of Deep Evidential Regressioncited by 40
  5. 2020Evaluating Robustness of Predictive Uncertainty Estimation: Are Dirichlet-based Models Reliable?cited by 39
  6. 2021Trustworthy Multimodal Regression with Mixture of Normal-inverse Gamma Distributionscited by 39
  7. 2023On Second-Order Scoring Rules for Epistemic Uncertainty Quantificationcited by 38
  8. 2023Second-Order Uncertainty Quantification: A Distance-Based Approachcited by 35
  9. 2020Regression Prior Networkscited by 28
  10. 2023Introducing an Improved Information-Theoretic Measure of Predictive Uncertaintycited by 28
  11. 2021Multivariate Deep Evidential Regressioncited by 23
  12. 2021Efficient and Robust LiDAR-Based End-to-End Navigationcited by 22
  13. 2023Second-Order Uncertainty Quantification: Variance-Based Measurescited by 19
  14. 2023Evidential Uncertainty Quantification: A Variance-Based Perspectivecited by 14
  15. 2021Improving evidential deep learning via multi-task learningcited by 13
  16. 2022Evidential Conditional Neural Processescited by 10
  17. 2023Dirichlet-based Uncertainty Quantification for Personalized Federated Learning with Improved Posterior Networkscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 5; 2025: 15; 2026: 15.

35 papers.

Sort:

By area:

  1. 2024Conformalized Credal Regions for Classification with Ambiguous Ground Truthcited by 19
  2. 2025Uncertainty Quantification for Machine Learning in Healthcare: A Surveycited by 14
  3. 2024Credal Wrapper of Model Averaging for Uncertainty Estimation in Classificationcited by 13
  4. 2025Uncertainty Estimation by Flexible Evidential Deep Learningcited by 10
  5. 2024Provable Uncertainty Decomposition via Higher-Order Calibrationcited by 9
  6. 2025Position: Epistemic uncertainty estimation methods are fundamentally incompletecited by 9
  7. 2025Credal and Interval Deep Evidential Classificationscited by 8
  8. 2025A calibration test for evaluating set-based epistemic uncertainty representationscited by 7
  9. 2026Learning Credal Ensembles via Distributionally Robust Optimizationcited by 7
  10. 2024Enhancing Uncertainty Quantification in Drug Discovery with Censored Regression Labelscited by 5
  11. 2025Reducing Aleatoric and Epistemic Uncertainty through Multi-modal Data Acquisitioncited by 5
  12. 2025Epistemic Wrapping for Uncertainty Quantificationcited by 5
23 more
  1. 2025Uncertainty Quantification for Regression using Proper Scoring Rulescited by 5
  2. 2026Uncertainty in Physics and AI: Taxonomy, Quantification, and Validationcited by 3
  3. 2025Rethinking Approximate Gaussian Inference in Classificationcited by 2
  4. 2025PRESCRIBE: Predicting Single-Cell Responses with Bayesian Estimationcited by 2
  5. 2025Deep Ensembles for Epistemic Uncertainty: A Frequentist Perspectivecited by 2
  6. 2025Generalized Regularized Evidential Deep Learning Models: Theory and Comprehensive Evaluationcited by 2
  7. 2026Evidential Transformation Network: Turning Pretrained Models into Evidential Models for Post-hoc Uncertainty Estimationcited by 2
  8. 2024Uncertainties of Latent Representations in Computer Visioncited by 1
  9. 2025On Uncertainty Calibration for Equivariant Functionscited by 1
  10. 2026A formal definition and meta-model for a machine theory of mindcited by 1
  11. 2026A Ground-Truth Framework for Uncertainty Disentanglement with Posterior Riskcited by 1
  12. 2025Principled Input-Output-Conditioned Post-Hoc Uncertainty Estimation for Regression Networkscited by 0
  13. 2025Foundations of Artificial Intelligence Frameworks: Notion and Limits of AGIcited by 0
  14. 2026Beyond a Single Signal: SPECTREG2, A Unified MultiExpert Anomaly Detector for Unknown Unknownscited by 0
  15. 2026Do Not Imitate, Reinforce: Iterative Classification via Belief Refinementcited by 0
  16. 2026Plug-in Losses for Evidential Deep Learning: A Simplified Framework for Uncertainty Estimation that Includes the Softmax Classifiercited by 0
  17. 2026Courtroom Analogy: New Perspective on Uncertainty-Aware Classificationcited by 0
  18. 2026Can we trust our models? Epistemic calibration in second-order classificationcited by 0
  19. 2026Uncertainty quantification for trustworthy deep learning: Methods and measurescited by 0
  20. 2026Uncertainty-Driven Replay Memory for Reinforcement Learningcited by 0
  21. 2026Lucid Dreaming for World Models: Learning to Doubt Imagination and Decide by Trustcited by 0
  22. 2026Beyond Answer Confidence: A Controlled Audit of Self-Knowledge in a Black-Box Decision Modelcited by 0
  23. 2026Rethinking Epistemic Uncertainty in Node Classification through Information Growthcited by 0