Learning graph representations is a fundamental task aimed at capturing
various properties of graphs in vector space. The most recent methods learn
such representations for static networks. However, real world networks evolve
over time and have varying dynamics. Capturing such evolutio