mlsanityData to 14 Oct 2026

Link Prediction with Untrained Message Passing Layers

arXiv 2406.16687 ↗First posted 24 Jun 2024cs.LG cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

Message passing neural networks (MPNNs) operate on graphs by exchanging information between neigbouring nodes. MPNNs have been successfully applied to various node-, edge-, and graph-level tasks in areas like molecular science, computer vision, natural language processing, and combinatorial optimization. However, most MPNNs require training on large amounts of labeled data, which can be costly and time-consuming. In this work, we explore the use of various untrained message passing layers in graph neural networks, i.e. variants of popular message passing architecture where we remove all trainable parameters that are used to transform node features in the message passing step. Focusing on link prediction, we find that untrained message passing layers can lead to competitive and even superior performance compared to fully trained MPNNs, especially in the presence of high-dimensional features. We provide a theoretical analysis of untrained message passing by relating the inner products of features implicitly produced by untrained message passing layers to path-based topological node similarity measures. As such, untrained message passing architectures can be viewed as a highly efficient and interpretable approach to link prediction.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 32 of its 57 references that are in the map. First posted on arXiv 24 Jun 2024.

Builds on

32 of 57 references.

References · 32

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2017Inductive Representation Learning on Large Graphscited by 6,320
  3. 2018How Powerful are Graph Neural Networks?cited by 4,393
  4. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  5. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,163
  6. 2019Simplifying Graph Convolutional Networkscited by 1,477
  7. 2016Variational Graph Auto-Encoderscited by 1,382
  8. 2016Revisiting Semi-Supervised Learning with Graph Embeddingscited by 976
  9. 2018Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networkscited by 881
  10. 2018Link Prediction Based on Graph Neural Networkscited by 832
  11. 2017Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Rankingcited by 335
  12. 2021Neural Bellman-Ford Networks: A General Graph Neural Network Framework for Link Predictioncited by 275
20 more
  1. 2019Revisiting Graph Neural Networks: All We Have is Low-Pass Filterscited by 261
  2. 2020The Surprising Power of Graph Neural Networks with Random Node Initializationcited by 198
  3. 2020Random Features Strengthen Graph Neural Networkscited by 186
  4. 2021Frame Averaging for Invariant and Equivariant Network Designcited by 168
  5. 2022Approximately Equivariant Networks for Imperfectly Symmetric Dynamicscited by 97
  6. 2022Graph Neural Networks for Link Prediction with Subgraph Sketchingcited by 81
  7. 2022Neo-GNNs: Neighborhood Overlap-aware Graph Neural Networks for Link Predictioncited by 78
  8. 2023Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarkingcited by 69
  9. 2023Neural Common Neighbor with Completion for Link Predictioncited by 64
  10. 2021On Positional and Structural Node Features for Graph Neural Networks on Non-attributed Graphscited by 49
  11. 2022OOD Link Prediction Generalization Capabilities of Message-Passing GNNs in Larger Test Graphscited by 38
  12. 2021Neural Link Prediction with Walk Poolingcited by 37
  13. 2023Fine-grained Expressivity of Graph Neural Networkscited by 25
  14. 2023Pure Message Passing Can Estimate Common Neighbor for Link Predictioncited by 18
  15. 2022You Can Have Better Graph Neural Networks by Not Training Weights at All: Finding Untrained GNNs Ticketscited by 14
  16. 2022Equivariant Mesh Attention Networkscited by 12
  17. 2021Learning Equivariant Energy Based Models with Equivariant Stein Variational Gradient Descentcited by 10
  18. 2019Initialization for Network Embedding: A Graph Partition Approachcited by 8
  19. 2024Training-free Graph Neural Networks and the Power of Labels as Featurescited by 5
  20. 2024You do not have to train Graph Neural Networks at all on text-attributed graphscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2026Fixed Aggregation Features Can Rival GNNscited by 4
  2. 2025Deep Graph Learning will stall without Network Sciencecited by 1
  3. 2026Do Temporal Link Predictors Need Learned Memory? A Smoothed-Count Baseline with a Handful of Parameterscited by 0