mlsanityData to 7 Oct 2026

Encoding Concepts in Graph Neural Networks

arXiv 2207.13586 ↗First posted 27 Jul 2022cs.LG cs.AI cs.LOcited by 18

Not on the map. Why?

Abstract (arXiv)

The opaque reasoning of Graph Neural Networks induces a lack of human trust. Existing graph network explainers attempt to address this issue by providing post-hoc explanations, however, they fail to make the model itself more interpretable. To fill this gap, we introduce the Concept Encoder Module, the first differentiable concept-discovery approach for graph networks. The proposed approach makes graph networks explainable by design by first discovering graph concepts and then using these to solve the task. Our results demonstrate that this approach allows graph networks to: (i) attain model accuracy comparable with their equivalent vanilla versions, (ii) discover meaningful concepts that achieve high concept completeness and purity scores, (iii) provide high-quality concept-based logic explanations for their prediction, and (iv) support effective interventions at test time: these can increase human trust as well as significantly improve model performance.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 31 of its 47 references that are in the map. First posted on arXiv 27 Jul 2022.

Builds on

31 of 47 references.

References · 31

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  4. 2018How Powerful are Graph Neural Networks?cited by 4,355
  5. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  6. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,414
  7. 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,085
  8. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  9. 2018Sanity Checks for Saliency Mapscited by 1,149
  10. 2017Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPRcited by 1,086
  11. 2020Concept Bottleneck Modelscited by 1,022
  12. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 735
19 more
  1. 2020TUDataset: A collection of benchmark datasets for learning with graphscited by 718
  2. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  3. 2016Interaction Networks for Learning about Objects, Relations and Physicscited by 503
  4. 2017Interpretation of Neural Networks is Fragilecited by 443
  5. 2020Parameterized Explainer for Graph Neural Networkcited by 423
  6. 2017The (Un)reliability of saliency methodscited by 331
  7. 2019Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networkscited by 241
  8. 2019On Completeness-aware Concept-Based Explanations in Deep Neural Networkscited by 239
  9. 2020Concept Whitening for Interpretable Image Recognitioncited by 225
  10. 2020PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networkscited by 210
  11. 2021CF-GNNExplainer: Counterfactual Explanations for Graph Neural Networkscited by 118
  12. 2020PinnerSage: Multi-Modal User Embedding Framework for Recommendations at Pinterestcited by 75
  13. 2021Robust Counterfactual Explanations on Graph Neural Networkscited by 72
  14. 2020Now You See Me (CME): Concept-based Model Extractioncited by 51
  15. 2021Entropy-based Logic Explanations of Neural Networkscited by 51
  16. 2021Logic Explained Networkscited by 45
  17. 2021GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networkscited by 35
  18. 2022Trust in AI: Interpretability is not necessary or sufficient, while black-box interaction is necessary and sufficientcited by 13
  19. 2020MEME: Generating RNN Model Explanations via Model Extractioncited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 8; 2024: 4; 2025: 0; 2026: 2.

18 papers.

Sort:

  1. 2022Global Explainability of GNNs via Logic Combination of Learned Conceptscited by 64
  2. 2023Interpretable Neural-Symbolic Concept Reasoningcited by 56
  3. 2023Learning to Receive Help: Intervention-Aware Concept Embedding Modelscited by 41
  4. 2022Explaining the Explainers in Graph Neural Networks: a Comparative Studycited by 34
  5. 2024Reconsidering Faithfulness in Regular, Self-Explainable and Domain Invariant GNNscited by 14
  6. 2022MEGAN: Multi-Explanation Graph Attention Networkcited by 7
  7. 2022GCondNet: A Novel Method for Improving Neural Networks on Small High-Dimensional Tabular Datacited by 6
  8. 2023SHARCS: Shared Concept Space for Explainable Multimodal Learningcited by 6
  9. 2023Interpretable Graph Networks Formulate Universal Algebra Conjecturescited by 4
  10. 2023Everybody Needs a Little HELP: Explaining Graphs via Hierarchical Conceptscited by 2
  11. 2024Global Concept Explanations for Graphs by Contrastive Learningcited by 2
  12. 2024Automated Molecular Concept Generation and Labeling with Large Language Modelscited by 2
6 more
  1. 2023Quantifying the Intrinsic Usefulness of Attributional Explanations for Graph Neural Networks with Artificial Simulatability Studiescited by 1
  2. 2023Digital Histopathology with Graph Neural Networks: Concepts and Explanations for Clinicianscited by 1
  3. 2024Disentangled and Self-Explainable Node Representation Learningcited by 1
  4. 2023GCI: A (G)raph (C)oncept (I)nterpretation Frameworkcited by 0
  5. 2026Combating Spurious Correlations in Graph Interpretability via Self-Reflectioncited by 0
  6. 2026Exploring Concept Subspace for Self-explainable Text-Attributed Graph Learningcited by 0