In the domain of heterogeneous graph representation learning, traditional methodologies often rely excessively on manually crafted meta-paths and neighbor aggregation mechanisms when faced with ...

A novel concept of quantifying graph non-isomorphism is introduced to measure structural differences between graphs, and thus overcoming the strict limitations of traditional graph isomorphism tests.

Classic Graph Convolutional Networks (GCNs) often learn node representation holistically, which would ignore the distinct impacts from different neighbors when aggregating their features to update a ...

Heterogeneous graphs organize data with nodes and edges, and have been widely used in various graph-centric applications. Often, some data are omitted during manual construction, leading to data ...