Heterogeneous Graph Structure Learning through the Lens of Data-generating Processes
This addresses the gap in graph machine learning for real-world applications with heterogeneous patterns, such as social networks or knowledge graphs, by providing a novel method for structure learning, though it is incremental as it extends existing homogeneous graph techniques to heterogeneous settings.
The paper tackles the problem of learning graph structure from observed data for heterogeneous graphs, where nodes and edges have multiple types, by introducing the first approach for heterogeneous graph structure learning (HGSL) and demonstrating its effectiveness in edge type identification and edge weight recovery through experiments on synthetic and real-world datasets.
Inferring the graph structure from observed data is a key task in graph machine learning to capture the intrinsic relationship between data entities. While significant advancements have been made in learning the structure of homogeneous graphs, many real-world graphs exhibit heterogeneous patterns where nodes and edges have multiple types. This paper fills this gap by introducing the first approach for heterogeneous graph structure learning (HGSL). To this end, we first propose a novel statistical model for the data-generating process (DGP) of heterogeneous graph data, namely hidden Markov networks for heterogeneous graphs (H2MN). Then we formalize HGSL as a maximum a-posterior estimation problem parameterized by such DGP and derive an alternating optimization method to obtain a solution together with a theoretical justification of the optimization conditions. Finally, we conduct extensive experiments on both synthetic and real-world datasets to demonstrate that our proposed method excels in learning structure on heterogeneous graphs in terms of edge type identification and edge weight recovery.