Corpus ID: 152282292

Universal Invariant and Equivariant Graph Neural Networks

@inproceedings{Keriven2019UniversalIA,
  title={Universal Invariant and Equivariant Graph Neural Networks},
  author={Nicolas Keriven and Gabriel Peyr{\'e}},
  booktitle={NeurIPS},
  year={2019}
}
  • Nicolas Keriven, Gabriel Peyré
  • Published in NeurIPS 2019
  • Computer Science, Mathematics
  • Graph Neural Networks (GNN) come in many flavors, but should always be either invariant (permutation of the nodes of the input graph does not affect the output) or equivariant (permutation of the input permutes the output. [...] Key Result Finally, unlike many previous settings that consider a fixed number of nodes, our results show that a GNN defined by a single set of parameters can approximate uniformly well a function defined on graphs of varying size.Expand Abstract

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 42 CITATIONS

    On Universal Equivariant Set Networks

    VIEW 1 EXCERPT
    CITES BACKGROUND

    A Simple Proof of the Universality of Invariant/Equivariant Graph Neural Networks

    VIEW 12 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Provably Powerful Graph Networks

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Neural Networks on Groups

    Permutation Invariant Graph Generation via Score-Based Generative Modeling

    VIEW 2 EXCERPTS
    CITES BACKGROUND

    FILTER CITATIONS BY YEAR

    2019
    2020

    CITATION STATISTICS

    • 4 Highly Influenced Citations

    • Averaged 21 Citations per year from 2019 through 2020

    • 80% Increase in citations per year in 2020 over 2019

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 39 REFERENCES

    Invariant and Equivariant Graph Networks

    VIEW 16 EXCERPTS
    HIGHLY INFLUENTIAL

    Universal approximations of invariant maps by neural networks

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    On the Universality of Invariant Networks

    VIEW 16 EXCERPTS
    HIGHLY INFLUENTIAL

    How Powerful are Graph Neural Networks?

    VIEW 2 EXCERPTS

    Symmetries and discriminability in feedforward network architectures

    VIEW 1 EXCERPT

    Group Equivariant Convolutional Networks

    VIEW 3 EXCERPTS