Fine-Grained Graph Generation through Latent Mixture Scheduling

TL;DR

This paper introduces a conditional variational autoencoder with a mixture scheduler for fine-grained, structure-aware graph generation, enabling precise control over generated graph properties.

Contribution

It proposes a novel mixture scheduling approach that dynamically aligns graph and property representations for improved fidelity and controllability in graph generation.

Findings

Achieves high-quality graph generation with strong property control.

Outperforms recent baselines on five real-world datasets.

Demonstrates effectiveness in applications like drug discovery and social networks.

Abstract

Structure aware graph generation aims to generate graphs that satisfy given topological properties. It has applications in domains such as drug discovery, social network modeling, and knowledge graph construction. Unlike existing methods that only provide coarse control over graph properties, we introduce a novel conditional variational autoencoder for fine-grained structural control in graph generation. The approach refines the decoder's latent space by dynamically aligning graph- and property-driven representations to improve both graph fidelity and control satisfaction. Specifically, the approach implements a mixture scheduler that progressively integrates graph and control priors. Experiments on five real-world datasets show the efficacy of the proposed model compared to recent baselines, achieving high generation quality while maintaining high controllability.