Abstract
Spatio-temporal graphs such as traffic networks or gene regulatory systemspresent challenges for the existing deep learning methods due to the complexityof structural changes over time. To address these issues, we introduceSpatio-Temporal Deep Graph Infomax (STDGI)---a fully unsupervised noderepresentation learning approach based on mutual information maximization thatexploits both the temporal and spatial dynamics of the graph. Our model tacklesthe challenging task of node-level regression by training embeddings tomaximize the mutual information between patches of the graph, at any given timestep, and between features of the central nodes of patches, in the future. Wedemonstrate through experiments and qualitative studies that the learnedrepresentations can successfully encode relevant information about the inputgraph and improve the predictive performance of spatio-temporal auto-regressiveforecasting models.