CoChat
Sign Up
Cited by 2,872Advanced Graph Neural NetworksMachine Learning and AlgorithmsGraph Theory and Algorithms

Deeper Insights Into Graph Convolutional Networks for Semi-Supervised Learning

Open access
Li, Qimai, Han, Zhichao, Wu, Xiao-ming·2018·Proceedings of the AAAI Conference on Artificial Intelligence
DOI 10.1609/aaai.v32i1.11604
Open this paper in CoChat — free account

Abstract

Many interesting problems in machine learning are being revisited with new deep learning tools. For graph-based semi-supervised learning, a recent important development is graph convolutional networks (GCNs), which nicely integrate local vertex features and graph topology in the convolutional layers. Although the GCN model compares favorably with other state-of-the-art methods, its mechanisms are not clear and it still requires considerable amount of labeled data for validation and model selection. In this paper, we develop deeper insights into the GCN model and address its fundamental limits. First, we show that the graph convolution of the GCN model is actually a special form of Laplacian smoothing, which is the key reason why GCNs work, but it also brings potential concerns of over-smoothing with many convolutional layers. Second, to overcome the limits of the GCN model with shallow architectures, we propose both co-training and self-training approaches to train GCNs. Our approaches significantly improve GCNs in learning with very few labels, and exempt them from requiring additional labels for validation. Extensive experiments on benchmarks have verified our theory and proposals.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1609/aaai.v32i1.11604

Cites

4 of 4 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Regular graph-based methods ignore the sequential nature of underlying texts.

    “However, regular graph-based methods ignore the sequential nature of underlying texts.”

    From Emotion Classification With Visibility Graphs (2025) · cites this paper as [20]

    The Experimental Setup (p. 6) describes documents as bag-of-words vectors recording word presence or absence, which do not capture word order. The paper does not establish the broader claim that regular graph-based methods generally ignore the sequential nature of texts. (LLM, conf=0.82)

    Experimental — accuracy not yet measured

  • GCNs primarily focus on local graph structures, which restricts their ability to utilize global context within graphs fully.

    “Furthermore, GCNs primarily focus on local graph structures, which restrict their ability to utilize global context within graphs fully [20].”

    From Emotion Classification With Visibility Graphs (2025) · cites this paper as [20]

  • Incorporating a significant number of graph convolution layers in network architectures can result in over-smoothing of dataset features, thereby diminishing the performance of the training process.

    “As for the GNN-based approaches, incorporating a significant number of graph convolution layers in network architectures can result in over-smoothing of dataset features, thereby diminishing the performance of the training process [20].”

    From Emotion Classification With Visibility Graphs (2025) · cites this paper as [20]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 29

  1. [Belkin, Niyogi, and Sindhwani] Belkin, M.; Niyogi, P.; and Sindhwani, V. 2006. Manifold regularization: A geometric framework for learning from labeled and unlabeled examples. Journal of machine learning research 7:2434.
  2. [Blum and Chawla] Blum, A., and Chawla, S. 2001. Learning from labeled and unlabeled data using graph mincuts. In ICML , 19-26. ACM.
  3. [Blum and Mitchell] Blum, A., and Mitchell, T. 1998. Combining labeled and unlabeled data with co-training. In Proceedings of the 11th annual conference on Computational learning theory , 92-100. ACM.
  4. [Blum et al.] Blum, A.; Lafferty, J.; Rwebangira, M.; and Reddy, R. 2004. Semi-supervised learning using randomized mincuts. In ICML , 13. ACM.
  5. [Bruna et al.] Bruna, J.; Zaremba, W.; Szlam, A.; and LeCun, Y. 2014. Spectral networks and locally connected networks on graphs. ICLR .
  6. [Chapelle and Zien] Chapelle, O., and Zien, A. 2005. Semisupervised classification by low density separation. In Proceedings of the Tenth International Workshop on Artificial Intelligence and Statistics , 57-64. Max-PlanckGesellschaft.
  7. [Defferrard, Bresson, and Vandergheynst] Defferrard, M.; Bresson, X.; and Vandergheynst, P. 2016. Convolutional neural networks on graphs with fast localized spectral filtering. In NIPS , 3844-3852. Curran Associates, Inc.
  8. [Glorot and Bengio] Glorot, X., and Bengio, Y. 2010. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics , 249256. PMLR.
  9. [Guo et al.] Guo, H.; Tang, R.; Ye, Y.; Li, Z.; and He, X. 2017. A graph-based push service platform. In International Conference on Database Systems for Advanced Applications , 636-648. Springer.
  10. D. K.; Vandergheynst, P.; and Gribonval, R. 2011. Wavelets on graphs via spectral graph theory. Applied and Computational Harmonic Analysis 30(2):129-150.
  11. [Joachims] Joachims, T. 2003. Transductive learning via spectral graph partitioning. In ICML , 290-297. ACM.
  12. [Kipf and Welling] Kipf, T. N., and Welling, M. 2017. Semisupervised classification with graph convolutional networks. In ICLR .
  13. [Lake, Salakhutdinov, and Tenenbaum] Lake, B. M.; Salakhutdinov, R.; and Tenenbaum, J. B. 2015. Humanlevel concept learning through probabilistic program induction. Science 350(6266):1332-1338.
  14. [Perozzi, Al-Rfou, and Skiena] Perozzi, B.; Al-Rfou, R.; and Skiena, S. 2014. Deepwalk: Online learning of social representations. In Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining , 701-710. ACM.
  15. [Rasmus et al.] Rasmus, A.; Berglund, M.; Honkala, M.; Valpola, H.; and Raiko, T. 2015. Semi-supervised learning with ladder networks. In NIPS , 3546-3554. Curran Associates, Inc.
  16. [Rezende et al.] Rezende, D.; Danihelka, I.; Gregor, K.; Wierstra, D.; et al. 2016. One-shot generalization in deep generative models. In ICML 2016 , 1521-1529. ACM.
  17. [Sandryhaila and Moura] Sandryhaila, A., and Moura, J. M. 2013. Discrete signal processing on graphs. IEEE transactions on signal processing 61(7):1644-1656.
  18. [Shuman et al.] Shuman, D. I.; Narang, S. K.; Frossard, P.; Ortega, A.; and Vandergheynst, P. 2013. The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains. IEEE Signal Processing Magazine 30(3):83-98.
  19. [Talukdar and Crammer] Talukdar, P. P., and Crammer, K. 2009. New regularized algorithms for transductive learning. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases , 442-457. Springer.
  20. [Taubin] Taubin, G. 1995. A signal processing approach to fair surface design. In Proceedings of the 22nd annual conference on Computer graphics and interactive techniques , 351-358. ACM.
  21. [Von Luxburg] Von Luxburg, U. 2007. A tutorial on spectral clustering. Statistics and computing 17(4):395-416.
  22. [Weston et al.] Weston, J.; Ratle, F.; Mobahi, H.; and Collobert, R. 2008. Deep learning via semi-supervised embedding. In Proceedings of the 25th international conference on Machine learning , 1168-1175. ACM.
  23. [Wu et al.] Wu, X.; Li, Z.; So, A. M.; Wright, J.; and Chang, S.-f. 2012. Learning with Partially Absorbing Random Walks. In NIPS 25 , 3077-3085. Curran Associates, Inc.
  24. [Wu, Li, and Chang] Wu, X.-M.; Li, Z.; and Chang, S. 2013. Analyzing the harmonic structure in graph-based learning. In NIPS .
  25. [Yang, Cohen, and Salakhutdinov] Yang, Z.; Cohen, W. W.; and Salakhutdinov, R. 2016. Revisiting semi-supervised learning with graph embeddings. In Proceedings of the 33rd International conference on Machine learning , 40-48. ACM.
  26. [Zachary] Zachary, W. W. 1977. An information flow model for conflict and fission in small groups. Journal of anthropological research 33(4):452-473.
  27. [Zhou et al.] Zhou, D.; Bousquet, O.; Lal, T. N.; Weston, J.; and Sch¨ olkopf, B. 2004. Learning with local and global consistency. In NIPS 16 , 321-328. MIT Press.
  28. [Zhu and Goldberg] Zhu, X., and Goldberg, A. 2009. Introduction to semi-supervised learning. Synthesis Lectures on Artificial Intelligence and Machine Learning 3(1):1-130.
  29. [Zhu, Ghahramani, and Lafferty] Zhu, X.; Ghahramani, Z.; and Lafferty, J. D. 2003. Semi-supervised learning using gaussian fields and harmonic functions. In ICML 2003 , 912919. ACM.

Cited by

Sorted by citations

Authors
Li, Q., Han, Z., & Wu, X.-M.
Affiliations
Hong Kong Polytechnic University (HK)
ORCID
Venue
Proceedings of the AAAI Conference on Artificial Intelligence
Publisher
Association for the Advancement of Artificial Intelligence (AAAI)
Published
2018-04-29
Volume
32
Issue
1
ISSN
  • 2159-5399 (ISSN-L)
  • 2374-3468
DOI
10.1609/aaai.v32i1.11604
OpenAlex
W2964051675
Cited by
2,872
References
4
Citations by year
2026: 226 · 2025: 380 · 2024: 454 · 2023: 491 · 2022: 446 · 2021: 466 · 2020: 289 · 2019: 112 · 2018: 8
FWCI
89.30
Type
conference paper
Language
en
Open access
Locations
  • Proceedings of the AAAI Conference on Artificial Intelligence (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) (submittedVersion)
  • arXiv (Cornell University)
Funders
Hong Kong Polytechnic University (1-ZVJJ)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Advanced Graph Neural Networks
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Algorithms
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Graph Theory and Algorithms
Keywords
graph convolutional networks, semi-supervised learning, graph convolution, Laplacian smoothing, over-smoothing, co-training, self-training, graph topology

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review