mlsanityData to 7 Oct 2026

TAM: Topology-Aware Margin Loss for Class-Imbalanced Node Classification

arXiv 2206.12917 ↗First posted 26 Jun 2022cs.LG cs.AIcited by 41

Not on the map. Why?

Abstract (arXiv)

Learning unbiased node representations under class-imbalanced graph data is challenging due to interactions between adjacent nodes. Existing studies have in common that they compensate the minor class nodes `as a group' according to their overall quantity (ignoring node connections in graph), which inevitably increase the false positive cases for major nodes. We hypothesize that the increase in these false positive cases is highly affected by the label distribution around each node and confirm it experimentally. In addition, in order to handle this issue, we propose Topology-Aware Margin (TAM) to reflect local topology on the learning objective. Our method compares the connectivity pattern of each node with the class-averaged counter-part and adaptively adjusts the margin accordingly based on that. Our method consistently exhibits superiority over the baselines on various node classification benchmark datasets with representative GNN architectures.

Key facts. Cited by 41 papers in the map (data through 7 Oct 2026). Builds on 32 of its 48 references that are in the map. First posted on arXiv 26 Jun 2022.

Builds on

32 of 48 references.

References · 32

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  3. 2017Graph Attention Networkscited by 8,854
  4. 2017Inductive Representation Learning on Large Graphscited by 6,284
  5. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
  6. 2018Graph Convolutional Neural Networks for Web-Scale Recommender Systemscited by 1,133
  7. 2016Revisiting Semi-Supervised Learning with Graph Embeddingscited by 970
  8. 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 821
  9. 2019Decoupling Representation and Classifier for Long-Tailed Recognitioncited by 806
  10. 2020Long-tail learning via logit adjustmentcited by 605
  11. 2018FastGCN: Fast Learning with Graph Convolutional Networks via Importance Samplingcited by 583
  12. 2019Multi-scale Attributed Node Embeddingcited by 543
20 more
  1. 2019MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixingcited by 442
  2. 2019BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognitioncited by 403
  3. 2020Balanced Meta-Softmax for Long-Tailed Visual Recognitioncited by 397
  4. 2021Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methodscited by 309
  5. 2020Equalization Loss for Long-Tailed Object Recognitioncited by 285
  6. 2020Long-tailed Recognition by Routing Diverse Distribution-Aware Expertscited by 270
  7. 2020Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effectcited by 256
  8. 2020Graph Neural Networks with Heterophilycited by 202
  9. 2020Learning From Multiple Experts: Self-paced Knowledge Distillation for Long-tailed Classificationcited by 175
  10. 2020Disentangling Label Distribution for Long-tailed Visual Recognitioncited by 172
  11. 2021GraphSMOTE: Imbalanced Node Classification on Graphs with Graph Neural Networkscited by 147
  12. 2020Seesaw Loss for Long-Tailed Instance Segmentationcited by 146
  13. 2021ACE: Ally Complementary Experts for Solving Long-Tailed Recognition in One-Shotcited by 86
  14. 2021RSG: A Simple but Effective Module for Learning Imbalanced Datasetscited by 61
  15. 2021Topology-Imbalance Learning for Semi-Supervised Node Classificationcited by 56
  16. 2021Adaptive Class Suppression Loss for Long-Tail Object Detectioncited by 55
  17. 2020Posterior Re-calibration for Imbalanced Datasetscited by 53
  18. 2021ImGAGN:Imbalanced Network Embedding via Generative Adversarial Graph Networkscited by 51
  19. 2020Class-Weighted Classification: Trade-offs and Robust Approachescited by 30
  20. 2020Network Embedding with Completely-imbalanced Labelscited by 21

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 13; 2024: 15; 2025: 7; 2026: 6.

41 papers.

Sort:

By area:

  1. 2024A Survey of Graph Neural Networks in Real world: Imbalance, Noise, Privacy and OOD Challengescited by 54
  2. 2023Class-Imbalanced Graph Learning without Class Rebalancingcited by 29
  3. 2023Class-Imbalanced Learning on Graphs: A Surveycited by 24
  4. 2023Hyperbolic Geometric Graph Representation Learning for Hierarchy-imbalance Node Classificationcited by 23
  5. 2024Partitioning Message Passing for Graph Fraud Detectioncited by 21
  6. 2023GraphSHA: Synthesizing Harder Samples for Class-Imbalanced Node Classificationcited by 17
  7. 2023A Survey of Imbalanced Learning on Graphs: Problems, Techniques, and Future Directionscited by 15
  8. 2024Federated Graph Learning with Structure Proxy Alignmentcited by 14
  9. 2023Rethinking Semi-Supervised Imbalanced Node Classification from Bias-Variance Decompositioncited by 13
  10. 2024Rethinking Fair Graph Neural Networks from Re-balancingcited by 11
  11. 2023Semantic-aware Node Synthesis for Imbalanced Heterogeneous Information Networkscited by 9
  12. 2024LOGIN: A Large Language Model Consulted Graph Neural Network Training Frameworkcited by 8
29 more
  1. 2023Over-Sampling Strategy in Feature Space for Graphs based Class-imbalanced Bot Detectioncited by 7
  2. 2023RAHNet: Retrieval Augmented Hybrid Network for Long-tailed Graph Classificationcited by 7
  3. 2024DIVE: Subgraph Disagreement for Graph Out-of-Distribution Generalizationcited by 7
  4. 2024Mitigating Label Noise on Graph via Topological Sample Selectioncited by 6
  5. 2024Cluster-guided Contrastive Class-imbalanced Graph Classificationcited by 6
  6. 2024Edge Classification on Graphs: New Directions in Topological Imbalancecited by 4
  7. 2025IceBerg: Debiased Self-Training for Class-Imbalanced Node Classificationcited by 4
  8. 2024Topology-Aware Dynamic Reweighting for Distribution Shifts on Graphcited by 3
  9. 2024ECGN: A Cluster-Aware Approach to Graph Neural Networks for Imbalanced Classificationcited by 2
  10. 2025Mitigating Message Imbalance in Fraud Detection with Dual-View Graph Representation Learningcited by 2
  11. 2026Re-understanding Graph Unlearning through Memorizationcited by 2
  12. 2026EAGLE: Edge-Aware Graph Learning for Proactive Delivery Delay Prediction in Smart Logistics Networkscited by 2
  13. 2023Using Active Learning Methods to Strategically Select Essays for Automated Scoringcited by 1
  14. 2023Heterophily-Based Graph Neural Network for Imbalanced Classificationcited by 1
  15. 2024Class-Balanced and Reinforced Active Learning on Graphscited by 1
  16. 2024Automated Loss function Search for Class-imbalanced Node Classificationcited by 1
  17. 2024SiMilarity-Enhanced Homophily for Multi-View Heterophilous Graph Clusteringcited by 1
  18. 2025NeuBM: Mitigating Model Bias in Graph Neural Networks through Neutral Input Calibrationcited by 1
  19. 2025GraphIFE: Rethinking Graph Imbalance Node Classification via Invariant Learningcited by 1
  20. 2023VIGraph: Generative Self-supervised Learning for Class-Imbalanced Node Classificationcited by 0
  21. 2023Understanding Community Bias Amplification in Graph Representation Learningcited by 0
  22. 2024BuffGraph: Enhancing Class-Imbalanced Node Classification via Buffer Nodescited by 0
  23. 2025UPL: Uncertainty-aware Pseudo-labeling for Imbalance Transductive Node Classificationcited by 0
  24. 2025Pure Node Selection for Imbalanced Graph Node Classificationcited by 0
  25. 2025GraphSB: Boosting Imbalanced Node Classification on Graphs through Structural Balancecited by 0
  26. 2026GraphSB: Boosting Imbalanced Node Classification on Graphs through Structural Balancecited by 0
  27. 2026NodeImport: Imbalanced Node Classification with Node Importance Assessmentcited by 0
  28. 2026OpenRTAG: A Comprehensive Benchmark for Robust Text-Attributed Graph Learning under Data Quality Degradationcited by 0
  29. 2026MDTE: Minority-Aware Diffusion over Temporal Edge Events for Imbalanced Node Classificationcited by 0