Spatio-Temporal Neural Networks for Space-Time Series Forecasting and Relations Discovery
This work addresses forecasting challenges in domains like epidemiology and traffic prediction, but it is incremental as it builds on existing neural network approaches with structured latent components.
The authors tackled the problem of forecasting time series with spatial dependencies by introducing a dynamical spatio-temporal recurrent neural network model, which achieved competitive performance compared to state-of-the-art baselines across epidemiology, geo-spatial statistics, and car-traffic prediction applications.
We introduce a dynamical spatio-temporal model formalized as a recurrent neural network for forecasting time series of spatial processes, i.e. series of observations sharing temporal and spatial dependencies. The model learns these dependencies through a structured latent dynamical component, while a decoder predicts the observations from the latent representations. We consider several variants of this model, corresponding to different prior hypothesis about the spatial relations between the series. The model is evaluated and compared to state-of-the-art baselines, on a variety of forecasting problems representative of different application areas: epidemiology, geo-spatial statistics and car-traffic prediction. Besides these evaluations, we also describe experiments showing the ability of this approach to extract relevant spatial relations.