mlsanityData to 7 Oct 2026

Robust Counterfactual Explanations on Graph Neural Networks

arXiv 2107.04086 ↗First posted 8 Jul 2021cs.LG cs.AIcited by 72

Not on the map. Why?

Abstract (arXiv)

Massive deployment of Graph Neural Networks (GNNs) in high-stake applications generates a strong demand for explanations that are robust to noise and align well with human intuition. Most existing methods generate explanations by identifying a subgraph of an input graph that has a strong correlation with the prediction. These explanations are not robust to noise because independently optimizing the correlation for a single input can easily overfit noise. Moreover, they do not align well with human intuition because removing an identified subgraph from an input graph does not necessarily change the prediction result. In this paper, we propose a novel method to generate robust counterfactual explanations on GNNs by explicitly modelling the common decision logic of GNNs on similar input graphs. Our explanations are naturally robust to noise because they are produced from the common decision boundaries of a GNN that govern the predictions of many similar input graphs. The explanations also align well with human intuition because removing the set of edges identified by an explanation from the input graph changes the prediction significantly. Exhaustive experiments on many public datasets demonstrate the superior performance of our method.

Key facts. Cited by 72 papers in the map (data through 7 Oct 2026). Builds on 33 of its 54 references that are in the map. First posted on arXiv 8 Jul 2021.

Builds on

33 of 54 references.

References · 33

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Graph Attention Networkscited by 8,854
  3. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  4. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  5. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  6. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  7. 2018Sanity Checks for Saliency Mapscited by 1,149
  8. 2018Link Prediction Based on Graph Neural Networkscited by 829
  9. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 735
  10. 2019Graph Neural Networks for Social Recommendationcited by 711
  11. 2017Interpretation of Neural Networks is Fragilecited by 443
  12. 2020Parameterized Explainer for Graph Neural Networkcited by 423
21 more
  1. 2018Adversarial Attacks on Neural Networks for Graph Datacited by 395
  2. 2014On the Number of Linear Regions of Deep Neural Networkscited by 378
  3. 2016Top-down Neural Attention by Excitation Backpropcited by 318
  4. 2021On Explainability of Graph Neural Networks via Subgraph Explorationscited by 280
  5. 2019Adversarial Attacks on Graph Neural Networks via Meta Learningcited by 261
  6. 2019Adversarial Attacks and Defenses in Images, Graphs and Text: A Reviewcited by 234
  7. 2017One-Shot Imitation Learningcited by 232
  8. 2020XGNN: Towards Model-Level Explanations of Graph Neural Networkscited by 211
  9. 2020PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networkscited by 210
  10. 2019Topology Attack and Defense for Graph Neural Networks: An Optimization Perspectivecited by 196
  11. 2013Maxout Networkscited by 186
  12. 2021CF-GNNExplainer: Counterfactual Explanations for Graph Neural Networkscited by 118
  13. 2020Causal Interpretability for Machine Learning -- Problems, Methods and Evaluationcited by 82
  14. 2021DIG: A Turnkey Library for Diving into Graph Deep Learning Researchcited by 76
  15. 2019Certifiable Robustness to Graph Perturbationscited by 62
  16. 2019Certifiable Robustness and Robust Training for Graph Convolutional Networkscited by 59
  17. 2013Submodular Optimization with Submodular Cover and Submodular Knapsack Constraintscited by 40
  18. 2021MEG: Generating Molecular Counterfactual Explanations for Deep Graph Networkscited by 32
  19. 2018Exact and Consistent Interpretation for Piecewise Linear Neural Networks: A Closed Form Solutioncited by 31
  20. 2017Programmable Agentscited by 14
  21. 2019Submodular Cost Submodular Cover with an Approximate Oraclecited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 15; 2023: 18; 2024: 14; 2025: 11; 2026: 11.

72 papers.

Sort:

By area:

  1. 2022Trustworthy Graph Neural Networks: Aspects, Methods and Trendscited by 88
  2. 2022GraphFramEx: Towards Systematic Evaluation of Explainability Methods for Graph Neural Networkscited by 57
  3. 2024A Survey of Graph Neural Networks in Real world: Imbalance, Noise, Privacy and OOD Challengescited by 54
  4. 2022CLEAR: Generative Counterfactual Explanations on Graphscited by 48
  5. 2022Global Counterfactual Explainer for Graph Neural Networkscited by 30
  6. 2022A Survey on Graph Counterfactual Explanations: Definitions, Methods, Evaluation, and Research Challengescited by 30
  7. 2023A Survey on Explainability of Graph Neural Networkscited by 28
  8. 2021Transport-based Counterfactual Modelscited by 27
  9. 2022A Survey of Explainable Graph Neural Networks: Taxonomy and Evaluation Metricscited by 26
  10. 2023Counterfactual Learning on Graphs: A Surveycited by 22
  11. 2022A Survey of Trustworthy Graph Learning: Reliability, Explainability, and Privacy Protectioncited by 20
  12. 2022FlowX: Towards Explainable Graph Neural Networks via Message Flowscited by 20
38 more
  1. 2023Provably Robust and Plausible Counterfactual Explanations for Neural Networks via Robust Optimisationcited by 20
  2. 2021A Framework and Benchmarking Study for Counterfactual Generating Methods on Tabular Datacited by 19
  3. 2024PowerGraph: A power grid benchmark dataset for graph neural networkscited by 19
  4. 2022Encoding Concepts in Graph Neural Networkscited by 18
  5. 2023Distill n' Explain: explaining graph neural networks using simple surrogatescited by 18
  6. 2024F-Fidelity: A Robust Framework for Faithfulness Evaluation of Explainable AIcited by 17
  7. 2023Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networkscited by 16
  8. 2022MotifExplainer: a Motif-based Graph Neural Network Explainercited by 11
  9. 2023GNNX-BENCH: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarkingcited by 11
  10. 2024GOAt: Explaining Graph Neural Networks via Graph Output Attributioncited by 10
  11. 2023GInX-Eval: Towards In-Distribution Evaluation of Graph Neural Network Explanationscited by 9
  12. 2023D4Explainer: In-Distribution GNN Explanations via Discrete Denoising Diffusioncited by 9
  13. 2022Cardinality-Minimal Explanations for Monotonic Neural Networkscited by 8
  14. 2022A.I. Robustness: a Human-Centered Perspective on Technological Challenges and Opportunitiescited by 8
  15. 2023Deep Backtracking Counterfactuals for Causally Compliant Explanationscited by 8
  16. 2024View-based Explanations for Graph Neural Networkscited by 8
  17. 2024TABCF: Counterfactual Explanations for Tabular Data Using a Transformer-Based VAEcited by 8
  18. 2023Robust Stochastic Graph Generator for Counterfactual Explanationscited by 7
  19. 2024Interval Abstractions for Robust Counterfactual Explanationscited by 7
  20. 2022GRETEL: A unified framework for Graph Counterfactual Explanation Evaluationcited by 6
  21. 2023Efficient GNN Explanation via Learning Removal-based Attributioncited by 6
  22. 2025Recent Advances in Malware Detection: Graph Learning and Explainabilitycited by 6
  23. 2025From Nodes to Narratives: Explaining Graph Neural Networks with LLMs and Graph Contextcited by 6
  24. 2024EiG-Search: Generating Edge-Induced Subgraphs for GNN Explanation in Linear Timecited by 5
  25. 2024Explainable Graph Neural Networks Under Firecited by 5
  26. 2021Towards the Explanation of Graph Neural Networks in Digital Pathology with Information Flowscited by 4
  27. 2022Privacy and Transparency in Graph Machine Learning: A Unified Perspectivecited by 4
  28. 2023Robust Ante-hoc Graph Explainer using Bilevel Optimizationcited by 4
  29. 2023When Graph Neural Network Meets Causality: Opportunities, Methodologies and An Outlookcited by 4
  30. 2024Generating Robust Counterfactual Witnesses for Graph Neural Networkscited by 4
  31. 2025B-XAIC Dataset: Benchmarking Explainable AI for Graph Neural Networks Using Chemical Datacited by 4
  32. 2024Motif-Consistent Counterfactuals with Adversarial Refinement for Graph-Level Anomaly Detectioncited by 3
  33. 2024Explaining Graph Neural Networks with Large Language Models: A Counterfactual Perspective for Molecular Property Predictioncited by 3
  34. 2022GANExplainer: GAN-based Graph Neural Networks Explainercited by 2
  35. 2025Graph Inverse Style Transfer for Counterfactual Explainabilitycited by 2
  36. 2026ATEX-CF: Attack-Informed Counterfactual Explanations for Graph Neural Networkscited by 2
  37. 2023Counterfactual Graph Transformer for Traffic Flow Predictioncited by 1
  38. 2023Adapting to Change: Robust Counterfactual Explanations in Dynamic Data Landscapescited by 1

50 most cited of 72 All 72 (JSON)