mlsanityData to 7 Oct 2026

Accurate and Scalable Estimation of Epistemic Uncertainty for Graph Neural Networks

arXiv 2309.10976 ↗First posted 20 Sep 2023cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Safe deployment of graph neural networks (GNNs) under distribution shift requires models to provide accurate confidence indicators (CI). However, while it is well-known in computer vision that CI quality diminishes under distribution shift, this behavior remains understudied for GNNs. Hence, we begin with a case study on CI calibration under controlled structural and feature distribution shifts and demonstrate that increased expressivity or model size do not always lead to improved CI performance. Consequently, we instead advocate for the use of epistemic uncertainty quantification (UQ) methods to modulate CIs. To this end, we propose G-DeltaUQ, a new single model UQ method that extends the recently proposed stochastic centering framework to support structured data and partial stochasticity. Evaluated across covariate, concept, and graph size shifts, G-DeltaUQ not only outperforms several popular UQ methods in obtaining calibrated CIs, but also outperforms alternatives when CIs are used for generalization gap prediction or OOD detection. Overall, our work not only introduces a new, flexible GNN UQ method, but also provides novel insights into GNN CIs on safety-critical tasks.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 63 of its 79 references that are in the map. First posted on arXiv 20 Sep 2023.

Builds on

63 of 79 references.

References · 63

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Graph Attention Networkscited by 8,854
  3. 2017On Calibration of Modern Neural Networkscited by 4,860
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  5. 2018How Powerful are Graph Neural Networks?cited by 4,355
  6. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  7. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  8. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  9. 2020Energy-based Out-of-distribution Detectioncited by 1,296
  10. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,274
  11. 2018Deep Anomaly Detection with Outlier Exposurecited by 939
  12. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 903
38 more
  1. 2018Link Prediction Based on Graph Neural Networkscited by 829
  2. 2020TUDataset: A collection of benchmark datasets for learning with graphscited by 718
  3. 2022Recipe for a General, Powerful, Scalable Graph Transformercited by 702
  4. 2020Benchmarking Graph Neural Networkscited by 672
  5. 2020On the Bottleneck of Graph Neural Networks and its Practical Implicationscited by 624
  6. 2020Bayesian Deep Learning and a Probabilistic Perspective of Generalizationcited by 480
  7. 2020Principal Neighbourhood Aggregation for Graph Netscited by 477
  8. 2021Understanding over-squashing and bottlenecks on graphs via curvaturecited by 471
  9. 2019Scaling Out-of-Distribution Detection for Real-World Settingscited by 465
  10. 2022ViM: Out-Of-Distribution with Virtual-logit Matchingcited by 404
  11. 2021Revisiting the Calibration of Modern Neural Networkscited by 387
  12. 2019Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibrationcited by 359
  13. 2021Graph Neural Networks with Learnable Structural and Positional Representationscited by 326
  14. 2019Verified Uncertainty Calibrationcited by 307
  15. 2018Graph R-CNN for Scene Graph Generationcited by 296
  16. 2015Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networkscited by 295
  17. 2021Unsolved Problems in ML Safetycited by 291
  18. 2022Long Range Graph Benchmarkcited by 258
  19. 2020Mix-n-Match: Ensemble and Compositional Methods for Uncertainty Calibration in Deep Learningcited by 190
  20. 2021A Fine-Grained Analysis on Distribution Shiftcited by 171
  21. 2019Understanding Attention and Generalization in Graph Neural Networkscited by 169
  22. 2021From Stars to Subgraphs: Uplifting Any GNN with Local Structure Awarenesscited by 154
  23. 2022Learning Causally Invariant Representations for Out-of-Distribution Generalization on Graphscited by 144
  24. 2022Leveraging Unlabeled Data to Predict Out-of-Distribution Performancecited by 141
  25. 2022GOOD: A Graph Out-of-Distribution Benchmarkcited by 140
  26. 2021PixMix: Dreamlike Pictures Comprehensively Improve Safety Measurescited by 131
  27. 2020Improving model calibration with accuracy versus uncertainty optimizationcited by 123
  28. 2020From Local Structures to Size Generalization in Graph Neural Networkscited by 119
  29. 2021Predicting with Confidence on Unseen Distributionscited by 116
  30. 2018Predicting the Generalization Gap in Deep Networks with Margin Distributionscited by 111
  31. 2020Calibration of Neural Networks using Splinescited by 97
  32. 2022Equivariant and Stable Positional Encoding for More Powerful Graph Neural Networkscited by 94
  33. 2022A Generalization of ViT/MLP-Mixer to Graphscited by 93
  34. 2021Be Confident! Towards Trustworthy Graph Neural Networks via Confidence Calibrationcited by 91
  35. 2018Bayesian graph convolutional neural networks for semi-supervised classificationcited by 88
  36. 2021Size-Invariant Graph Representations for Graph Classification Extrapolationscited by 82
  37. 2020Bayesian Graph Neural Networks with Adaptive Connection Samplingcited by 81
  38. 2023Attending to Graph Transformerscited by 81

50 most cited of 63 All 63 (JSON)

Built on by

Not cited yet.