mlsanityData to 7 Oct 2026

HC-GST: Heterophily-aware Distribution Consistency based Graph Self-training

arXiv 2407.17787 ↗First posted 25 Jul 2024cs.SI cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Graph self-training (GST), which selects and assigns pseudo-labels to unlabeled nodes, is popular for tackling label sparsity in graphs. However, recent study on homophily graphs show that GST methods could introduce and amplify distribution shift between training and test nodes as they tend to assign pseudo-labels to nodes they are good at. As GNNs typically perform better on homophilic nodes, there could be potential shifts towards homophilic pseudo-nodes, which is underexplored. Our preliminary experiments on heterophilic graphs verify that these methods can cause shifts in homophily ratio distributions, leading to textittraining bias that improves performance on homophilic nodes while degrading it on heterophilic ones. Therefore, we study a novel problem of reducing homophily ratio distribution shifts during self-training on heterophilic graphs. A key challenge is the accurate calculation of homophily ratios and their distributions without extensive labeled data. To tackle them, we propose a novel Heterophily-aware Distribution Consistency-based Graph Self-Training (HC-GST) framework, which estimates homophily ratios using soft labels and optimizes a selection vector to align pseudo-nodes with the global homophily ratio distribution. Extensive experiments on both homophilic and heterophilic graphs show that HC-GST effectively reduces training bias and enhances self-training performance.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 32 of its 46 references that are in the map. First posted on arXiv 25 Jul 2024.

Builds on

32 of 46 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. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  4. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  5. 2019Simplifying Graph Convolutional Networkscited by 1,470
  6. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,099
  7. 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 821
  8. 2016Deep Transfer Learning with Joint Adaptation Networkscited by 609
  9. 2020Adaptive Universal Generalized PageRank Graph Neural Networkcited by 592
  10. 2018FastGCN: Fast Learning with Graph Convolutional Networks via Importance Samplingcited by 583
  11. 2020Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networkscited by 563
  12. 2019Multi-scale Attributed Node Embeddingcited by 543
20 more
  1. 2019MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixingcited by 442
  2. 2019Composition-based Multi-Relational Graph Convolutional Networkscited by 395
  3. 2020Rethinking Pre-training and Self-trainingcited by 352
  4. 2017Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Rankingcited by 332
  5. 2021Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methodscited by 309
  6. 2022Graph Neural Networks for Graphs with Heterophily: A Surveycited by 175
  7. 2019Multi-Stage Self-Supervised Learning for Graph Convolutional Networks on Graphs with Few Labelscited by 111
  8. 2022Out-Of-Distribution Generalization on Graphs: A Surveycited by 97
  9. 2021Be Confident! Towards Trustworthy Graph Neural Networks via Confidence Calibrationcited by 91
  10. 2020Graph Prototypical Networks for Few-shot Learning on Attributed Networkscited by 90
  11. 2021Shift-Robust GNNs: Overcoming the Limitations of Localized Graph Training Datacited by 86
  12. 2022Debiased Self-Training for Semi-Supervised Learningcited by 62
  13. 2021Simplifying approach to Node Classification in Graph Neural Networkscited by 56
  14. 2021Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutionscited by 38
  15. 2022Confidence May Cheat: Self-Training on Graph Neural Networks under Distribution Shiftcited by 38
  16. 2021Block Modeling-Guided Graph Convolutional Neural Networkscited by 32
  17. 2017Robust Unsupervised Domain Adaptation for Neural Networks via Moment Alignmentcited by 29
  18. 2022FAITH: Few-Shot Graph Classification with Hierarchical Task Graphscited by 22
  19. 2023Few-shot Node Classification with Extremely Weak Supervisioncited by 14
  20. 2024Distribution Consistency based Self-Training for Graph Neural Networks with Sparse Labelscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 1.

2 papers.

  1. 2024A Survey of Deep Graph Learning under Distribution Shifts: from Graph Out-of-Distribution Generalization to Adaptationcited by 16
  2. 2026Heterophily-Aware Adaptive Knowledge Distillation for Hypergraph Neural Networkscited by 0