Structured Sequence Modeling with Graph Convolutional Recurrent Networks

Youngjoo Seo, Michaël Defferrard, Pierre Vandergheynst, Xavier Bresson,
International Conference on Neural Information Processing (ICONIP), 2017.

This paper introduces Graph Convolutional Recurrent Network (GCRN), a deep learning model able to predict structured sequences of data. Precisely, GCRN is a generalization of classical recurrent neural networks (RNN) to data structured by an arbitrary graph. Such structured sequences can represent series of frames in videos, spatio-temporal measurements on a network of sensors, or random walks on a vocabulary graph for natural language modeling. The proposed model combines convolutional neural networks (CNN) on graphs to identify spatial structures and RNN to find dynamic patterns. We study two possible architectures of GCRN, and apply the models to two practical problems: predicting moving MNIST data, and modeling natural language with the Penn Treebank dataset. Experiments show that exploiting simultaneously graph spatial and dynamic information about data can improve both precision and learning speed.

@inproceedings{gcrn,
  title = {Structured Sequence Modeling with Graph Convolutional Recurrent Networks},
  author = {Seo, Youngjoo and Defferrard, Micha\"el and Vandergheynst, Pierre and Bresson, Xavier},
  booktitle = {International Conference on Neural Information Processing (ICONIP)},
  year = {2017},
  archiveprefix = {arXiv},
  eprint = {1612.07659},
  url = {https://arxiv.org/abs/1612.07659},
}

Resources

PDF available at arXiv, OpenReview, ICONIP, EPFL.

Related: code, slides.

Compilation

Compile the latex source into a PDF with make. Run make clean to remove temporary files and make arxiv.zip to prepare an archive to be uploaded on arXiv.

Figures

All the figures are in the figures folder. PDFs can be generated with make figures.