Graph-Time Convolutional Neural Networks
This work addresses spatiotemporal data analysis for domains like sensor networks or traffic prediction, but it is incremental as it builds on existing graph convolutional methods.
The authors tackled the challenge of learning representations from spatiotemporal data represented as graphs by developing a Graph-Time Convolutional Neural Network (GTCNN) that uses parametric product graphs and zero-pad pooling, achieving competitive results in experiments with synthetic and real data.
Spatiotemporal data can be represented as a process over a graph, which captures their spatial relationships either explicitly or implicitly. How to leverage such a structure for learning representations is one of the key challenges when working with graphs. In this paper, we represent the spatiotemporal relationships through product graphs and develop a first principle graph-time convolutional neural network (GTCNN). The GTCNN is a compositional architecture with each layer comprising a graph-time convolutional module, a graph-time pooling module, and a nonlinearity. We develop a graph-time convolutional filter by following the shift-and-sum principles of the convolutional operator to learn higher-level features over the product graph. The product graph itself is parametric so that we can learn also the spatiotemporal coupling from data. We develop a zero-pad pooling that preserves the spatial graph (the prior about the data) while reducing the number of active nodes and the parameters. Experimental results with synthetic and real data corroborate the different components and compare with baseline and state-of-the-art solutions.