mlsanityData to 7 Oct 2026

Prototypical Graph Contrastive Learning

arXiv 2106.09645 ↗First posted 17 Jun 2021cs.LG cs.AIcited by 37

Not on the map. Why?

Abstract (arXiv)

Graph-level representations are critical in various real-world applications, such as predicting the properties of molecules. But in practice, precise graph annotations are generally very expensive and time-consuming. To address this issue, graph contrastive learning constructs instance discrimination task which pulls together positive pairs (augmentation pairs of the same graph) and pushes away negative pairs (augmentation pairs of different graphs) for unsupervised representation learning. However, since for a query, its negatives are uniformly sampled from all graphs, existing methods suffer from the critical sampling bias issue, i.e., the negatives likely having the same semantic structure with the query, leading to performance degradation. To mitigate this sampling bias issue, in this paper, we propose a Prototypical Graph Contrastive Learning (PGCL) approach. Specifically, PGCL models the underlying semantic structure of the graph data via clustering semantically similar graphs into the same group, and simultaneously encourages the clustering consistency for different augmentations of the same graph. Then given a query, it performs negative sampling via drawing the graphs from those clusters that differ from the cluster of query, which ensures the semantic difference between query and its negative samples. Moreover, for a query, PGCL further reweights its negative samples based on the distance between their prototypes (cluster centroids) and the query prototype such that those negatives having moderate prototype distance enjoy relatively large weights. This reweighting strategy is proved to be more effective than uniform sampling. Experimental results on various graph benchmarks testify the advantages of our PGCL over state-of-the-art methods. Code is publicly available at https://github.com/ha-lins/PGCL.

Key facts. Cited by 37 papers in the map (data through 7 Oct 2026). Builds on 51 of its 96 references that are in the map. First posted on arXiv 17 Jun 2021.

Builds on

51 of 96 references.

References · 51

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  3. 2017Graph Attention Networkscited by 8,854
  4. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  5. 2017Inductive Representation Learning on Large Graphscited by 6,284
  6. 2018How Powerful are Graph Neural Networks?cited by 4,355
  7. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  8. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  9. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,845
  10. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  11. 2016Variational Graph Auto-Encoderscited by 1,373
  12. 2018Learning deep representations by mutual information estimation and maximizationcited by 1,265
38 more
  1. 2020Graph Contrastive Learning with Augmentationscited by 1,203
  2. 2018Deep Graph Infomaxcited by 1,101
  3. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,099
  4. 2018Deep Clustering for Unsupervised Learning of Visual Featurescited by 1,020
  5. 2019Strategies for Pre-training Graph Neural Networkscited by 964
  6. 2018Hierarchical Graph Representation Learning with Differentiable Poolingcited by 786
  7. 2017Focal Loss for Dense Object Detectioncited by 749
  8. 2020TUDataset: A collection of benchmark datasets for learning with graphscited by 718
  9. 2020Contrastive Multi-View Representation Learning on Graphscited by 651
  10. 2020Prototypical Contrastive Learning of Unsupervised Representationscited by 636
  11. 2020Contrastive Learning with Hard Negative Samplescited by 619
  12. 2020Deep Graph Contrastive Representation Learningcited by 535
  13. 2019A Theoretical Analysis of Contrastive Unsupervised Representation Learningcited by 530
  14. 2019Self-labelling via simultaneous clustering and representation learningcited by 525
  15. 2020Graph Contrastive Learning with Adaptive Augmentationcited by 519
  16. 2020GCC: Graph Contrastive Coding for Graph Neural Network Pre-Trainingcited by 461
  17. 2019InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximizationcited by 441
  18. 2020Debiased Contrastive Learningcited by 354
  19. 2019On Mutual Information Maximization for Representation Learningcited by 315
  20. 2020GPT-GNN: Generative Pre-Training of Graph Neural Networkscited by 302
  21. 2021Graph Self-Supervised Learning: A Surveycited by 289
  22. 2021Graph Contrastive Learning Automatedcited by 287
  23. 2017graph2vec: Learning Distributed Representations of Graphscited by 228
  24. 2019Graph Matching Networks for Learning the Similarity of Graph Structured Objectscited by 202
  25. 2021Self-supervised Learning on Graphs: Contrastive, Generative,or Predictivecited by 136
  26. 2021Self-supervised Graph-level Representation Learning with Local and Global Structurecited by 125
  27. 2020Sub-graph Contrast for Scalable Self-Supervised Graph Representation Learningcited by 90
  28. 2019Unsupervised Deep Learning by Neighbourhood Discoverycited by 78
  29. 2021SUGAR: Subgraph Neural Network with Reinforcement Pooling and Self-Supervised Mutual Information Mechanismcited by 75
  30. 2020Contrastive Self-supervised Learning for Graph Classificationcited by 68
  31. 2019Heterogeneous Deep Graph Infomaxcited by 42
  32. 2021A Primer on Contrastive Pretraining in Language Processing: Methods, Lessons Learned and Perspectivescited by 41
  33. 2020Convolutional Kernel Networks for Graph-Structured Datacited by 38
  34. 2016The Multiscale Laplacian Graph Kernelcited by 31
  35. 2020Self-supervised Training of Graph Convolutional Networkscited by 24
  36. 2019Spatio-Temporal Deep Graph Infomaxcited by 21
  37. 2020Iterative Graph Self-Distillationcited by 21
  38. 2014Accelerating Minibatch Stochastic Gradient Descent using Stratified Samplingcited by 20

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 14; 2023: 8; 2024: 5; 2025: 6; 2026: 2.

37 papers.

Sort:

By area:

  1. 2021Augmentation-Free Self-Supervised Learning on Graphscited by 94
  2. 2022Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learningcited by 86
  3. 2022Mugs: A Multi-Granular Self-Supervised Learning Frameworkcited by 54
  4. 2021Graph Communal Contrastive Learningcited by 34
  5. 2022Bringing Your Own View: Graph Contrastive Learning without Prefabricated Data Augmentationscited by 28
  6. 2022Disentangled Contrastive Learning for Social Recommendationcited by 27
  7. 2023Self-Supervised Temporal Graph learning with Temporal and Structural Intensity Alignmentcited by 25
  8. 2022Evaluating Self-Supervised Learning for Molecular Graph Embeddingscited by 22
  9. 2022Towards Understanding Why Mask-Reconstruction Pretraining Helps in Downstream Taskscited by 18
  10. 2022Towards Faithful and Consistent Explanations for Graph Neural Networkscited by 13
  11. 2023Spatial-Temporal Graph Learning with Adversarial Contrastive Adaptationcited by 13
  12. 2022TopoImb: Toward Topology-level Imbalance in Learning from Graphscited by 12
25 more
  1. 2022Relational Self-Supervised Learning on Graphscited by 11
  2. 2022Toward Robust Graph Semi-Supervised Learning against Extreme Data Scarcitycited by 11
  3. 2022Link Prediction with Non-Contrastive Learningcited by 11
  4. 2023Affinity Uncertainty-based Hard Negative Mining in Graph Contrastive Learningcited by 9
  5. 2022Variational Graph Generator for Multi-View Graph Clusteringcited by 7
  6. 2023Faithful and Consistent Graph Neural Network Explanations with Rationale Alignmentcited by 6
  7. 2023Hierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learningcited by 6
  8. 2024Impact of Domain Knowledge and Multi-Modality on Intelligent Molecular Property Prediction: A Systematic Surveycited by 6
  9. 2022ProtoCLIP: Prototypical Contrastive Language Image Pretrainingcited by 5
  10. 2025Rethinking Graph Contrastive Learning through Relative Similarity Preservationcited by 5
  11. 2025Building Egocentric Procedural AI Assistant: Methods, Benchmarks, and Challengescited by 5
  12. 2022Improving Molecular Pretraining with Complementary Featurizationscited by 4
  13. 2024Unifying Graph Contrastive Learning via Graph Message Augmentationcited by 2
  14. 2023Topology-aware Debiased Self-supervised Graph Learning for Recommendationcited by 1
  15. 2024ExGRG: Explicitly-Generated Relation Graph for Self-Supervised Representation Learningcited by 1
  16. 2024Topology Reorganized Graph Contrastive Learning with Mitigating Semantic Driftcited by 1
  17. 2025Graph Self-Supervised Learning with Learnable Structural and Positional Encodingscited by 1
  18. 2025Exploiting Text Semantics for Few and Zero Shot Node Classification on Text-attributed Graphcited by 1
  19. 2025Clustering-Oriented Generative Attribute Graph Imputationcited by 1
  20. 2023Capturing Fine-grained Semantics in Contrastive Graph Representation Learningcited by 0
  21. 2023Gaussian Graph with Prototypical Contrastive Learning in E-Commerce Bundle Recommendationcited by 0
  22. 2024Multi-Task Curriculum Graph Contrastive Learning with Clustering Entropy Guidancecited by 0
  23. 2025Negative Metric Learning for Graphscited by 0
  24. 2026ChemHyperMag: Physics-informed magnetic hypergraph learning improves molecular ADMET predictioncited by 0
  25. 2026Rethinking Data Augmentation under Covariate Shift: Invariant-Guided Diffusion and Prototype Reweightingcited by 0