2018

Graph-Based Deep Modeling and Real Time Forecasting of Sparse Spatio-Temporal Data

Wang, Bao, Luo, Xiyang, Zhang, Fangbo et al.

Understand

We present a generic framework for spatio-temporal (ST) data modeling, analysis, and forecasting, with a special focus on data that is sparse in both space and time.

Our multi-scaled framework is a seamless coupling of two major components: a self-exciting point process that models the macroscale statistical behaviors of the ST data and a graph structured recurrent neural network (GSRNN) to discover the microscale patterns of the ST data on the inferred graph.
This novel deep neural network (DNN) incorporates the real time interactions of the graph nodes to enable more accurate real time forecasting.
The effectiveness of our method is demonstrated on both crime and traffic forecasting.

Built on

Approximations by superpositions of sigmoidal functions
G. Cybenko · 1989
Earlier work this paper cites.
Long short-term memory
S. Hochreiter and J. Schmidhuber · 1997
Earlier work this paper cites.
Estimation of space-time branching process models in seismology using an em-type algorithm
A. Veen and F. P. Schoenberg · 2008
Earlier work this paper cites.
The split bregman method for l1-regularized problems
T. Goldstein and S.J. Osher · 2009
Earlier work this paper cites.
Qgis geographic information system
QGIS Development Team · 2009
Earlier work this paper cites.
Learning social infectivity in sparse low-rank networks using multi-dimensional hawkes processes
K. Zhou, H.Y. Zha, and L. Song · 2013
Earlier work this paper cites.
Adam: A method for stochastic optimization
D. Kingma and J. Ba · 2015
Earlier work this paper cites.

Similar

Hawkes processes
Original
P. J. Laub, T. Taimre, and P. K. Pollett · 2015
Cited alongside, same era.
Deep learning
Y. LeCun, Y. Bengion, and G. Hinton · 2015
Cited alongside, same era.
Randomized controlled field trials of predictive policing
G. Mohler, M. Short, S. Malinowski, M. Johnson, G.E. Tita, A.L. Bertozzi, and P. J. Brantingham · 2015
Cited alongside, same era.
Recurrent marked temporal point process
N. Du, H. Dai, R. Trivedi, U. Upadhyay, M. Rodriguez, and L. Song · 2016
Cited alongside, same era.
Deep residual learning for image recognition
K. He, X. Zhang, S. Ren, and J. Sun · 2016
Cited alongside, same era.
Structural-rnn: Deep learning on spatio-temporal graphs
A. Jain, A. R. Zamir, S. Savarese, and A. Saxena · 2016
Cited alongside, same era.
Gated graph sequence neural network
Y. Li, R. Zemel, M. Brockschmidt, and D. Tarlow · 2016
Cited alongside, same era.

Then

Point-process models of social network interactions: parameter estimation and missing data recovery
J. R. Zipkin, F.P. Schoenberg, K. Coronges, and A. L. Bertozzi · 2016
Later among the works it cites.
Phase-functioned neural networks for character control
D. Holden, T. Komura, and J. Saito · 2017
Later among the works it cites.
Prediction of crime occurrence from multi-modal data using deep learning
H. Kang and H. Kang · 2017
Later among the works it cites.
Deep learning for real time crime forecasting and its ternization
Original
B. Wang, P. Yin, A. L. Bertozzi, P. J. Brantingham, S. J. Osher, and J. Xin · 2017
Later among the works it cites.
Deep learning for real time crime forecasting
B. Wang, D. Zhang, D. Zhang, P. J. Brantingham, and A. L. Bertozzi · 2017
Later among the works it cites.
Deep spatio-temporal residual networks for citywide crowd flows prediction
J. Zhang, Y. Zheng, and D. Qi · 2017
Later among the works it cites.
The concentration dynamics tradeoff in crime hot spotting
G. O. Mohler, M. B. Short, and P. J. Brantingham · 2018
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…