A Study on Graph-Structured Recurrent Neural Networks and Sparsification with Application to Epidemic Forecasting

  • Zhijian Li
  • Xiyang Luo
  • Bao Wang
  • Andrea L. Bertozzi
  • Jack XinEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 991)


We study epidemic forecasting on real-world health data by a graph-structured recurrent neural network (GSRNN). We achieve state-of-the-art forecasting accuracy on the benchmark CDC dataset. To improve model efficiency, we sparsify the network weights via a transformed-\(\ell _1\) penalty without losing prediction accuracy in numerical experiments.


Spatio-temporal data Spatio-temporal graph Graph structured recurrent neural network Epidemic forecasting Sparsification 



This material is based on research sponsored by the Air Force Research Laboratory and DARPA under agreement number FA8750-18-2-0066; the U.S. Department of Energy, Office of Science, DOE-SC0013838; the National Science Foundation DMS-1554564 (STROBE), DMS-1737770, DMS-1522383, IIS-1632935. The authors thank Profs. M. Hyman, and J. Lega for helpful discussions.


  1. 1.
  2. 2.
    Dinh, T., Xin, J.: Convergence of a relaxed variable splitting method for learning sparse neural networks via \(\ell _1, \ell _0\), and transformed-\(\ell _1\) penalties (2018). ArXiv: 1812.05719
  3. 3.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  4. 4.
    Jain, A., Zamir, A., Savarese, S., Saxena, A.: Structural-RNN: deep learning on spatio-temporal graphs. In: Conference on Computer Vision and Pattern Recognition (CVPR 2016) (2016)Google Scholar
  5. 5.
    Nsoesie, E., Brownstein, J., Ramakrishnan, N., Marathe, M.: A systematic review of studies on forecasting the dynamics of influenza outbreaks. Influenza Other Respir. Viruses 8(3), 309–316 (2014)CrossRefGoogle Scholar
  6. 6.
    Osher, S., Wang, B., Yin, P., Luo, X., Pham, M., Lin, A.: Laplacian smoothing gradient descent (2018). ArXiv:1806.06317
  7. 7.
    Pascanu, R., Mikolov, T., Bengio, Y.: On the difficulty of training recurrent neural networks. In: Proceedings of the 30th International Conference on Machine Learning (2013)Google Scholar
  8. 8.
    Perra, N., Goncalves, B.: Modeling and predicting human infectious disease. In: Social Phenomena, pp. 59–83. Springer (2015)Google Scholar
  9. 9.
    Volkova, S., Ayton, E., Porterfield, K., Corley, C.: Forecasting influenza-like illness dynamics for military populations using neural networks and social media. PLOS one 12(12), e0188941 (2017)CrossRefGoogle Scholar
  10. 10.
    Wang, B., Luo, X., Zhang, F., Yuan, B., Bertozzi, A., Brantingham, P.: Graph-based deep modeling and real time forecasting of sparse spatio-temporal data (2018). arXiv:1804.00684
  11. 11.
    Wang, B., Yin, P., Bertozzi, A., Brantingham, P., Osher, S., Xin, J.: Deep learning for real-time crime forecasting and its ternarization (2017). arXiv:1711.08833
  12. 12.
    Wang, B., Zhang, D., Zhang, D., Brantingham, P., Bertozzi, A.: Deep learning for real-time crime forecasting (2017). arXiv:1707.03340
  13. 13.
    Yang, S., Santillana, M., Kou, S.: Accurate estimation of influenza epidemics using Google search data via ARGO. Proc. Natl. Acad. Sci. 112(47) (2015)CrossRefGoogle Scholar
  14. 14.
    Zhang, S., Xin, J.: Minimization of transformed \(\ell _1\) penalty: theory, difference of convex function algorithm, and robust application in compressed sensing. Math. Program. Ser. B 169(1), 307–336 (2018)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Zhijian Li
    • 1
  • Xiyang Luo
    • 2
  • Bao Wang
    • 2
  • Andrea L. Bertozzi
    • 2
  • Jack Xin
    • 1
    Email author
  1. 1.UC IrvineIrvineUSA
  2. 2.UCLALos AngelesUSA

Personalised recommendations