Scalable Hierarchical Graph Generation via Soft Community Structure
We present Schema, which recursively decomposes a reference graph into a hierarchy of soft communities, assigning each node a membership distribution.
ProofPaper ↗
Key points
- Generating large attributed graphs requires reproducing the topology, generating attributes jointly with the structure, and remaining scalable.
- Many real-world graphs exist as a single large graph, so a generative model has to generalize from the one graph it is fit on, without independent samples.
- Generation is then split into three stages, each trained independently: (1) synthesizing node attributes conditioned on soft memberships, (2) generating intra-community edges from local structural context, and (3) modeling inter-community connections over bridge nodes whose membership mass is distributed across several communities.
- We also introduce an evaluation protocol that covers structural fidelity, memorization, downstream utility, and scalability.
Sources (1)
- [1]Scalable Hierarchical Graph Generation via Soft Community StructurearXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 03:39 PM
We present Schema, which recursively decomposes a reference graph into a hierarchy of soft communities, assigning each node a membership distribution.
Generating large attributed graphs requires reproducing the topology, generating attributes jointly with the structure, and remaining scalable.
Extractive summary: sentences quoted from the sources.