Temporal Graph Neural Networks for Irregular Data
This addresses forecasting problems in domains like traffic and climate modeling where data is irregularly sampled, representing an incremental improvement by integrating graph neural networks with time-continuous dynamics.
The paper tackles forecasting of graph-structured irregularly observed time series by proposing a temporal graph neural network model that handles irregular time steps and partial graph observations, achieving validation on simulated and real-world traffic and climate data.
This paper proposes a temporal graph neural network model for forecasting of graph-structured irregularly observed time series. Our TGNN4I model is designed to handle both irregular time steps and partial observations of the graph. This is achieved by introducing a time-continuous latent state in each node, following a linear Ordinary Differential Equation (ODE) defined by the output of a Gated Recurrent Unit (GRU). The ODE has an explicit solution as a combination of exponential decay and periodic dynamics. Observations in the graph neighborhood are taken into account by integrating graph neural network layers in both the GRU state update and predictive model. The time-continuous dynamics additionally enable the model to make predictions at arbitrary time steps. We propose a loss function that leverages this and allows for training the model for forecasting over different time horizons. Experiments on simulated data and real-world data from traffic and climate modeling validate the usefulness of both the graph structure and time-continuous dynamics in settings with irregular observations.