mlsanityData to 7 Oct 2026

On Positional and Structural Node Features for Graph Neural Networks on Non-attributed Graphs

arXiv 2107.01495 ↗First posted 3 Jul 2021cs.LG cs.AI cs.SIcited by 49

Not on the map. Why?

Abstract (arXiv)

Graph neural networks (GNNs) have been widely used in various graph-related problems such as node classification and graph classification, where superior performance is mainly established when natural node features are available. However, it is not well understood how GNNs work without natural node features, especially regarding the various ways to construct artificial ones. In this paper, we point out the two types of artificial node features, i.e., positional and structural node features, and provide insights on why each of them is more appropriate for certain tasks, i.e., positional node classification, structural node classification, and graph classification. Extensive experimental results on 10 benchmark datasets validate our insights, thus leading to a practical guideline on the choices between different artificial node features for GNNs on non-attributed graphs. The code is available at https://github.com/zjzijielu/gnn-positional-structural-node-features.

Key facts. Cited by 49 papers in the map (data through 7 Oct 2026). Builds on 32 of its 68 references that are in the map. First posted on arXiv 3 Jul 2021.

Builds on

32 of 68 references.

References · 32

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Inductive Representation Learning on Large Graphscited by 6,284
  3. 2018How Powerful are Graph Neural Networks?cited by 4,355
  4. 2014DeepWalk: Online Learning of Social Representationscited by 1,543
  5. 2016Variational Graph Auto-Encoderscited by 1,373
  6. 2018Link Prediction Based on Graph Neural Networkscited by 829
  7. 2020TUDataset: A collection of benchmark datasets for learning with graphscited by 718
  8. 2017Representation Learning on Graphs: Methods and Applicationscited by 456
  9. 2019A Fair Comparison of Graph Neural Networks for Graph Classificationcited by 244
  10. 2017struc2vec: Learning Node Representations from Structural Identitycited by 214
  11. 2019Position-aware Graph Neural Networkscited by 203
  12. 2020Combining Label Propagation and Simple Models Out-performs Graph Neural Networkscited by 199
20 more
  1. 2017Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2veccited by 197
  2. 2021Identity-aware Graph Neural Networkscited by 197
  3. 2020The Surprising Power of Graph Neural Networks with Random Node Initializationcited by 190
  4. 2020Random Features Strengthen Graph Neural Networkscited by 184
  5. 2020Design Space for Graph Neural Networkscited by 182
  6. 2020Graph Information Bottleneckcited by 168
  7. 2020Machine Learning on Graphs: A Model and Comprehensive Taxonomycited by 156
  8. 2020Rethinking pooling in graph neural networkscited by 73
  9. 2020Graph Policy Network for Transferable Active Learning on Graphscited by 49
  10. 2020Graph Convolutional Networks for Graphs Containing Missing Featurescited by 46
  11. 2020Learning on Attribute-Missing Graphscited by 42
  12. 2020Path Integral Based Convolution and Pooling for Graph Neural Networkscited by 35
  13. 2020Graph Cross Networks with Vertex Infomax Poolingcited by 31
  14. 2020Optimization of Graph Neural Networks with Natural Gradient Descentcited by 26
  15. 2019On Node Features for Graph Neural Networkscited by 20
  16. 2018A simple yet effective baseline for non-attributed graph classificationcited by 16
  17. 2018Understanding Regularized Spectral Clustering via Graph Conductancecited by 14
  18. 2018Contextual Graph Markov Model: A Deep and Generative Approach to Graph Processingcited by 13
  19. 2019Relation Learning on Social Networks with Multi-Modal Graph Edge Variational Autoencoderscited by 13
  20. 2021Graph Entropy Guided Node Embedding Dimension Selection for Graph Neural Networkscited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 9; 2023: 8; 2024: 12; 2025: 11; 2026: 4.

49 papers.

Sort:

  1. 2021Graph Neural Networks with Learnable Structural and Positional Representationscited by 326
  2. 2022Brain Network Transformercited by 92
  3. 2022BrainGB: A Benchmark for Brain Network Analysis with Graph Neural Networkscited by 73
  4. 2021On the Unreasonable Effectiveness of Feature propagation in Learning on Graphs with Missing Node Featurescited by 62
  5. 2022Interpretable Graph Neural Networks for Connectome-Based Brain Disorder Analysiscited by 58
  6. 2022GraphGDP: Generative Diffusion Processes for Permutation Invariant Graph Generationcited by 43
  7. 2023NeuroGraph: Benchmarks for Graph Machine Learning in Brain Connectomicscited by 33
  8. 2022Graph Auto-Encoder Via Neighborhood Wasserstein Reconstructioncited by 28
  9. 2022Data-Efficient Brain Connectome Analysis via Multi-Task Meta-Learningcited by 15
  10. 2021BrainNNExplainer: An Interpretable Graph Neural Network Framework for Brain Network based Disease Analysiscited by 13
  11. 2022Accurate Node Feature Estimation with Structured Variational Graph Autoencodercited by 12
  12. 2023TouchUp-G: Improving Feature Representation through Graph-Centric Finetuningcited by 12
37 more
  1. 2023PTGB: Pre-Train Graph Neural Networks for Brain Network Analysiscited by 10
  2. 2026Learning Posterior Predictive Distributions for Node Classification from Synthetic Graph Priorscited by 10
  3. 2022Graph Neural Network-based Android Malware Classification with Jumping Knowledgecited by 9
  4. 2024Feature Distribution on Graph Topology Mediates the Effect of Graph Convolution: Homophily Perspectivecited by 9
  5. 2021Graph Neural Networks for Graph Drawingcited by 8
  6. 2022How Can Graph Neural Networks Help Document Retrieval: A Case Study on CORD19 with Concept Map Generationcited by 7
  7. 2023Recovering Missing Node Features with Local Structure-based Embeddingscited by 5
  8. 2023Out of Sight, Still in Mind: Reasoning and Planning about Unobserved Objects with Video Tracking Enabled Memory Modelscited by 5
  9. 2025GRAIN: Exact Graph Reconstruction from Gradientscited by 5
  10. 2024Rethinking the Effectiveness of Graph Classification Datasets in Benchmarks for Assessing GNNscited by 4
  11. 2024Enhancing GNNs Performance on Combinatorial Optimization by Recurrent Feature Updatecited by 4
  12. 2024IOHunter: Graph Foundation Model to Uncover Online Information Operationscited by 4
  13. 2025View Space: Learning Representation across Arbitrary Graphscited by 4
  14. 2021Deep Fraud Detection on Non-attributed Graphcited by 3
  15. 2024Decoupled Subgraph Federated Learningcited by 3
  16. 2024Link Prediction with Untrained Message Passing Layerscited by 3
  17. 2025Incomplete Graph Learning: A Comprehensive Surveycited by 3
  18. 2023GRAN is superior to GraphRNN: node orderings, kernel- and graph embeddings-based metrics for graph generatorscited by 2
  19. 2024BrainMAP: Learning Multiple Activation Pathways in Brain Networkscited by 2
  20. 2023Creating generalizable downstream graph models with random projectionscited by 1
  21. 2024GNN Applied to Ego-nets for Friend Suggestionscited by 1
  22. 2025Simple yet Effective Node Property Prediction on Edge Streams under Distribution Shiftscited by 1
  23. 2025Urban Incident Prediction with Graph Neural Networks: Integrating Government Ratings and Crowdsourced Reportscited by 1
  24. 2025Bridging RDF Knowledge Graphs with Graph Neural Networks for Semantically-Rich Recommender Systemscited by 1
  25. 2026Structurally Human, Semantically Biased: Detecting LLM-Generated References with Embeddings and GNNscited by 1
  26. 2023Observe Locally, Classify Globally: Using GNNs to Identify Sparse Matrix Structurecited by 0
  27. 2024Imbalanced Graph-Level Anomaly Detection via Counterfactual Augmentation and Feature Learningcited by 0
  28. 2024Optimizing Federated Graph Learning with Inherent Structural Knowledge and Dual-Densely Connected GNNscited by 0
  29. 2024PropEnc: A Property Encoder for Graph Neural Networkscited by 0
  30. 2024Enhancing Future Link Prediction in Quantum Computing Semantic Networks through LLM-Initiated Node Featurescited by 0
  31. 2025Visualization Tasks for Unlabeled Graphscited by 0
  32. 2025How do Probabilistic Graphical Models and Graph Neural Networks Look at Network Data?cited by 0
  33. 2025Forecasting Faculty Placement from Patterns in Co-authorship Networkscited by 0
  34. 2025Learning Network Dismantling Without Handcrafted Inputscited by 0
  35. 2025Spectral Graph Neural Networks for Cognitive Task Classification in fMRI Connectomescited by 0
  36. 2026RAwR: Role-Aware Rewiring via Approximate Equitable Partitioncited by 0
  37. 2026Diffusion-Induced Spatial Attention Overlapping Community Detectioncited by 0