Corpus ID: 202783932

Diffusion Improves Graph Learning

@inproceedings{Klicpera2019DiffusionIG,
  title={Diffusion Improves Graph Learning},
  author={Johannes Klicpera and Stefan Weissenberger and Stephan G{\"u}nnemann},
  booktitle={NeurIPS},
  year={2019}
}
  • Johannes Klicpera, Stefan Weissenberger, Stephan Günnemann
  • Published in NeurIPS 2019
  • Computer Science, Mathematics
  • Graph convolution is the core of most Graph Neural Networks (GNNs) and usually approximated by message passing between direct (one-hop) neighbors. In this work, we remove the restriction of using only the direct neighbors by introducing a powerful, yet spatially localized graph convolution: Graph diffusion convolution (GDC). GDC leverages generalized graph diffusion, examples of which are the heat kernel and personalized PageRank. It alleviates the problem of noisy and often arbitrarily defined… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-4 OF 4 CITATIONS

    SIGN: Scalable Inception Graph Neural Networks

    VIEW 6 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    Adaptive Propagation Graph Convolutional Network

    VIEW 4 EXCERPTS
    CITES BACKGROUND
    HIGHLY INFLUENCED

    Continuous Graph Neural Networks

    VIEW 1 EXCERPT
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 103 REFERENCES

    Semi-Supervised Classification with Graph Convolutional Networks

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    A tutorial on spectral clustering

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Deep Graph Infomax

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Graph Attention Networks

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Stochastic blockmodels and community structure in networks

    • Brian Karrer, Mark E. J. Newman
    • Mathematics, Physics, Computer Science, Medicine
    • Physical review. E, Statistical, nonlinear, and soft matter physics
    • 2011
    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    On Spectral Clustering: Analysis and an algorithm

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Matrix Perturbation Theory

    VIEW 2 EXCERPTS
    HIGHLY INFLUENTIAL