mlsanityData to 7 Oct 2026

Improving Subgraph Recognition with Variational Graph Information Bottleneck

arXiv 2112.09899 ↗First posted 18 Dec 2021cs.LG cs.AIcited by 45

Not on the map. Why?

Abstract (arXiv)

Subgraph recognition aims at discovering a compressed substructure of a graph that is most informative to the graph property. It can be formulated by optimizing Graph Information Bottleneck (GIB) with a mutual information estimator. However, GIB suffers from training instability and degenerated results due to its intrinsic optimization process. To tackle these issues, we reformulate the subgraph recognition problem into two steps: graph perturbation and subgraph selection, leading to a novel Variational Graph Information Bottleneck (VGIB) framework. VGIB first employs the noise injection to modulate the information flow from the input graph to the perturbed graph. Then, the perturbed graph is encouraged to be informative to the graph property. VGIB further obtains the desired subgraph by filtering out the noise in the perturbed graph. With the customized noise prior for each input, the VGIB objective is endowed with a tractable variational upper bound, leading to a superior empirical performance as well as theoretical properties. Extensive experiments on graph interpretation, explainability of Graph Neural Networks, and graph classification show that VGIB finds better subgraphs than existing methods. Code is avaliable at https://github.com/Samyu0304/VGIB

Key facts. Cited by 45 papers in the map (data through 7 Oct 2026). Builds on 39 of its 60 references that are in the map. First posted on arXiv 18 Dec 2021.

Builds on

39 of 60 references.

References · 39

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Graph Attention Networkscited by 8,854
  3. 2017Inductive Representation Learning on Large Graphscited by 6,284
  4. 2018How Powerful are Graph Neural Networks?cited by 4,355
  5. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  7. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
  8. 2016Deep Variational Information Bottleneckcited by 915
  9. 2020What Makes for Good Views for Contrastive Learning?cited by 843
  10. 2018Hierarchical Graph Representation Learning with Differentiable Poolingcited by 786
  11. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 735
  12. 2018MINE: Mutual Information Neural Estimationcited by 680
27 more
  1. 2019DropEdge: Towards Deep Graph Convolutional Networks on Node Classificationcited by 679
  2. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 647
  3. 2018FastGCN: Fast Learning with Graph Convolutional Networks via Importance Samplingcited by 583
  4. 2019DeepGCNs: Can GCNs Go as Deep as CNNs?cited by 527
  5. 2019InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximizationcited by 441
  6. 2020Parameterized Explainer for Graph Neural Networkcited by 423
  7. 2021On Explainability of Graph Neural Networks via Subgraph Explorationscited by 280
  8. 2017Concrete Dropoutcited by 227
  9. 2018Adaptive Sampling Towards Fast Graph Representation Learningcited by 180
  10. 2020Graph Information Bottleneckcited by 168
  11. 2019ASAP: Adaptive Structure Aware Pooling for Learning Hierarchical Graph Representationscited by 156
  12. 2020Interpreting Graph Neural Networks for NLP With Differentiable Edge Maskingcited by 152
  13. 2016Information Dropout: Learning Optimal Representations Through Noisy Computationcited by 131
  14. 2020Graph Information Bottleneck for Subgraph Recognitioncited by 119
  15. 2020Restricting the Flow: Information Bottlenecks for Attributioncited by 110
  16. 2018Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flowcited by 103
  17. 2019Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneckcited by 101
  18. 2020Unsupervised Speech Decomposition via Triple Information Bottleneckcited by 87
  19. 2019InfoBot: Transfer and Exploration via the Information Bottleneckcited by 78
  20. 2019Significance-aware Information Bottleneck for Domain Adaptive Semantic Segmentationcited by 74
  21. 2020Subgraph Neural Networkscited by 73
  22. 2019Learning Efficient Multi-agent Communication: An Information Bottleneck Approachcited by 60
  23. 2019Semi-Supervised Graph Classification: A Hierarchical Graph Perspectivecited by 55
  24. 2020Graph Cross Networks with Vertex Infomax Poolingcited by 31
  25. 2021Recognizing Predictive Substructures with Subgraph Information Bottleneckcited by 28
  26. 2021Drop-Bottleneck: Learning Discrete Compressed Representation for Noise-Robust Explorationcited by 15
  27. 2019Efficient Algorithms for Densest Subgraph Discoverycited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 5; 2023: 16; 2024: 8; 2025: 8; 2026: 8.

45 papers.

Sort:

By area:

  1. 2023Towards Self-Interpretable Graph-Level Anomaly Detectioncited by 41
  2. 2023Mind the Label Shift of Augmentation-based Graph OOD Generalizationcited by 34
  3. 2023A Survey on Explainability of Graph Neural Networkscited by 28
  4. 2023Conditional Graph Information Bottleneck for Molecular Relational Learningcited by 24
  5. 2023Interpretable Prototype-based Graph Information Bottleneckcited by 23
  6. 2024Contrastive Graph Representation Learning with Adversarial Cross-view Reconstruction and Information Bottleneckcited by 21
  7. 2022A Survey of Trustworthy Graph Learning: Reliability, Explainability, and Privacy Protectioncited by 20
  8. 2022FlowX: Towards Explainable Graph Neural Networks via Message Flowscited by 20
  9. 2023Drug Discovery with Dynamic Goal-aware Fragmentscited by 19
  10. 2025A Comprehensive Survey on Self-Interpretable Neural Networkscited by 18
  11. 2022BrainIB: Interpretable Brain Network-based Psychiatric Diagnosis with Graph Information Bottleneckcited by 17
  12. 2023CI-GNN: A Granger Causality-Inspired Graph Neural Network for Interpretable Brain Network-Based Psychiatric Diagnosiscited by 16
33 more
  1. 2023Individual and Structural Graph Information Bottlenecks for Out-of-Distribution Generalizationcited by 15
  2. 2024Reconsidering Faithfulness in Regular, Self-Explainable and Domain Invariant GNNscited by 14
  3. 2024How Interpretable Are Interpretable Graph Neural Networks?cited by 13
  4. 2023Identifying Semantic Component for Robust Molecular Property Predictioncited by 10
  5. 2023Learning High-Order Relationships of Brain Regionscited by 10
  6. 2025Beyond Topological Self-Explainable GNNs: A Formal Explainability Perspectivecited by 10
  7. 2022Finding Diverse and Predictable Subgraphs for Graph Domain Generalizationcited by 9
  8. 2024Self-Explainable Temporal Graph Networks based on Graph Information Bottleneckcited by 9
  9. 2025Can Knowledge-Graph-based Retrieval Augmented Generation Really Retrieve What You Need?cited by 8
  10. 2022Deep Hypergraph Structure Learningcited by 7
  11. 2023Adversarially Robust Neural Architecture Search for Graph Neural Networkscited by 7
  12. 2025Recent Advances in Malware Detection: Graph Learning and Explainabilitycited by 6
  13. 2023How Faithful are Self-Explainable GNNs?cited by 5
  14. 2023Label Information Bottleneck for Label Enhancementcited by 4
  15. 2023Rumor Detection with Diverse Counterfactual Evidencecited by 4
  16. 2024Graph Augmentation for Recommendationcited by 4
  17. 2025B-XAIC Dataset: Benchmarking Explainable AI for Graph Neural Networks Using Chemical Datacited by 4
  18. 2023Graph Anomaly Detection at Group Level: A Topology Pattern Enhanced Unsupervised Approachcited by 3
  19. 2026GNN Explanations that do not Explain and How to find Themcited by 3
  20. 2026IBCircuit: Towards Holistic Circuit Discovery with Information Bottleneckcited by 3
  21. 2024Towards Understanding Sensitive and Decisive Patterns in Explainable AI: A Case Study of Model Interpretation in Geometric Deep Learningcited by 1
  22. 2024GaGSL: Global-augmented Graph Structure Learning via Graph Information Bottleneckcited by 1
  23. 2026Revealing Combinatorial Reasoning of GNNs via Graph Concept Bottleneck Layercited by 1
  24. 2023Structure-Sensitive Graph Dictionary Embedding for Graph Classificationcited by 0
  25. 2024GINTRIP: Interpretable Temporal Graph Regression using Information bottleneck and Prototype-based methodcited by 0
  26. 2025Interpretable High-order Knowledge Graph Neural Network for Predicting Synthetic Lethality in Human Cancerscited by 0
  27. 2025Soft causal learning for generalized molecule property prediction: An environment perspectivecited by 0
  28. 2025ARM-Explainer -- Explaining and improving graph neural network predictions for the maximum clique problem using node features and association rule miningcited by 0
  29. 2026Super Agents and Confounders: Influence of surrounding agents on vehicle trajectory predictioncited by 0
  30. 2026Exploring Concept Subspace for Self-explainable Text-Attributed Graph Learningcited by 0
  31. 2026Subgraph Concept Networks: Concept Levels in Graph Classificationcited by 0
  32. 2026Why Self-Inconsistency Arises in GNN Explanations and How to Exploit Itcited by 0
  33. 2026AIMing for Standardised Explainability Evaluation in GNNs: A Framework and Case Study on Graph Kernel Networkscited by 0