mlsanityData to 7 Oct 2026

Towards neural networks that provably know when they don't know

arXiv 1909.12180 ↗First posted 26 Sep 2019cs.LG cs.CV stat.MLcited by 69

Not on the map. Why?

Abstract (arXiv)

It has recently been shown that ReLU networks produce arbitrarily over-confident predictions far away from the training data. Thus, ReLU networks do not know when they don't know. However, this is a highly important property in safety critical applications. In the context of out-of-distribution detection (OOD) there have been a number of proposals to mitigate this problem but none of them are able to make any mathematical guarantees. In this paper we propose a new approach to OOD which overcomes both problems. Our approach can be used with ReLU networks and provides provably low confidence predictions far away from the training data as well as the first certificates for low confidence predictions in a neighborhood of an out-distribution point. In the experiments we show that state-of-the-art methods fail in this worst-case setting whereas our model can guarantee its performance while retaining state-of-the-art OOD performance.

Key facts. Cited by 69 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 Sep 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 19; 2022: 18; 2023: 14; 2024: 10; 2025: 3; 2026: 5.

69 papers.

Sort:

By area:

  1. 2021Generalized Out-of-Distribution Detection: A Surveycited by 796
  2. 2022Out-of-Distribution Detection with Deep Nearest Neighborscited by 596
  3. 2021DICE: Leveraging Sparsification for Out-of-Distribution Detectioncited by 166
  4. 2021SegmentMeIfYouCan: A Benchmark for Anomaly Segmentationcited by 150
  5. 2023Non-Parametric Outlier Synthesiscited by 129
  6. 2022POEM: Out-of-Distribution Detection with Posterior Samplingcited by 108
  7. 2022The Unreasonable Effectiveness of Random Pruning: Return of the Most Naive Baseline for Sparse Trainingcited by 85
  8. 2023Provable Dynamic Fusion for Low-Quality Multimodal Datacited by 81
  9. 2021Graph Posterior Network: Bayesian Predictive Uncertainty for Node Classificationcited by 79
  10. 2023Dream the Impossible: Outlier Imagination with Diffusion Modelscited by 77
  11. 2021SmoothMix: Training Confidence-calibrated Smoothed Classifiers for Certified Robustnesscited by 49
  12. 2022A Simple Approach to Improve Single-Model Deep Uncertainty via Distance-Awarenesscited by 49
38 more
  1. 2022Accountability in an Algorithmic Society: Relationality, Responsibility, and Robustness in Machine Learningcited by 47
  2. 2021Deep Ensembling with No Overhead for either Training or Testing: The All-Round Blessings of Dynamic Sparsitycited by 44
  3. 2022Partial and Asymmetric Contrastive Learning for Out-of-Distribution Detection in Long-Tailed Recognitioncited by 44
  4. 2023ID-like Prompt Learning for Few-Shot Out-of-Distribution Detectioncited by 42
  5. 2021How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Reviewcited by 36
  6. 2024How Does Unlabeled Data Provably Help Out-of-Distribution Detection?cited by 36
  7. 2021Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safetycited by 35
  8. 2022iDECODe: In-distribution Equivariance for Conformal Out-of-distribution Detectioncited by 27
  9. 2022Disentangling Epistemic and Aleatoric Uncertainty in Reinforcement Learningcited by 27
  10. 2021Uncertainty-Aware Reliable Text Classificationcited by 23
  11. 2022Breaking Down Out-of-Distribution Detection: Many Methods Based on OOD Training Data Estimate a Combination of the Same Core Quantitiescited by 23
  12. 2021Taxonomy of Machine Learning Safety: A Survey and Primercited by 22
  13. 2022Two Video Data Sets for Tracking and Retrieval of Out of Distribution Objectscited by 19
  14. 2021Provably Robust Detection of Out-of-distribution Data (almost) for freecited by 17
  15. 2023DOS: Diverse Outlier Sampling for Out-of-Distribution Detectioncited by 16
  16. 2023EAT: Towards Long-Tailed Out-of-Distribution Detectioncited by 16
  17. 2021Being a Bit Frequentist Improves Bayesian Neural Networkscited by 15
  18. 2023How to Overcome Curse-of-Dimensionality for Out-of-Distribution Detection?cited by 15
  19. 2021CODEs: Chamfer Out-of-Distribution Examples against Overconfidence Issuecited by 14
  20. 2021Natural Posterior Network: Deep Bayesian Uncertainty for Exponential Family Distributionscited by 13
  21. 2023Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shiftscited by 12
  22. 2023AUTO: Adaptive Outlier Optimization for Test-Time OOD Detectioncited by 12
  23. 2021SoK: Machine Learning Governancecited by 9
  24. 2024Forte : Finding Outliers with Representation Typicality Estimationcited by 9
  25. 2021Responsible and Regulatory Conform Machine Learning for Medicine: A Survey of Challenges and Solutionscited by 7
  26. 2022Test-Time Adaptation for Visual Document Understandingcited by 6
  27. 2024Out-of-Distribution Data: An Acquaintance of Adversarial Examples -- A Surveycited by 6
  28. 2024When and How Does In-Distribution Label Help Out-of-Distribution Detection?cited by 6
  29. 2024Out-Of-Distribution Detection with Diversification (Provably)cited by 6
  30. 2023DeepLens: Interactive Out-of-distribution Data Detection in NLP Modelscited by 5
  31. 2023Diffusion Denoised Smoothing for Certified and Adversarial Robust Out-Of-Distribution Detectioncited by 5
  32. 2021Task-Driven Detection of Distribution Shifts with Statistical Guarantees for Robot Learningcited by 4
  33. 2021Explanatory Pluralism in Explainable AIcited by 4
  34. 2022Detecting and Learning the Unknown in Semantic Segmentationcited by 4
  35. 2022On Attacking Out-Domain Uncertainty Estimation in Deep Neural Networkscited by 4
  36. 2024Your Classifier Can Be Secretly a Likelihood-Based OOD Detectorcited by 4
  37. 2022Explanatory Learning: Beyond Empiricism in Neural Networkscited by 3
  38. 2024Hyperbolic Metric Learning for Visual Outlier Detectioncited by 3

50 most cited of 69 All 69 (JSON)