Corpus ID: 219636268

Weisfeiler-Lehman Embedding for Molecular Graph Neural Networks

@article{Ishiguro2020WeisfeilerLehmanEF,
  title={Weisfeiler-Lehman Embedding for Molecular Graph Neural Networks},
  author={Katsuhiko Ishiguro and Kenta Oono and K. Hayashi},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.06909}
}
  • Katsuhiko Ishiguro, Kenta Oono, K. Hayashi
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • A graph neural network (GNN) is a good choice for predicting the chemical properties of molecules. Compared with other deep networks, however, the current performance of a GNN is limited owing to the “curse of depth.” Inspired by longestablished feature engineering in the field of chemistry, we expanded an atom representation using Weisfeiler-Lehman (WL) embedding, which is designed to capture local atomic patterns dominating the chemical properties of a molecule. In terms of representability… CONTINUE READING

    Figures and Tables from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 54 REFERENCES
    Adam: A Method for Stochastic Optimization
    • 49,675
    • PDF
    Deep Residual Learning for Image Recognition
    • 50,349
    • PDF
    Semi-Supervised Classification with Graph Convolutional Networks
    • 4,619
    • Highly Influential
    • PDF
    Distributed Representations of Words and Phrases and their Compositionality
    • 18,859
    • PDF
    Random Forests
    • 47,147
    • PDF
    Inductive Representation Learning on Large Graphs
    • 2,032
    • PDF
    Gated Graph Sequence Neural Networks
    • 1,075
    • PDF
    Neural Message Passing for Quantum Chemistry
    • 1,331
    • PDF