Combinatorial Complex Score-based Diffusion Modelling through Stochastic Differential Equations

TL;DR

This paper introduces a unified stochastic differential equation framework for generating complex combinatorial structures like graphs and hypergraphs, advancing generative AI capabilities beyond traditional graph generation.

Contribution

It generalizes score-based generative models to combinatorial complexes, unifying existing approaches and enabling the generation of higher-order structures.

Findings

Framework successfully generates complex objects like graphs and hypergraphs.

Competitively performs against state-of-the-art methods in graph and molecule generation.

Overcomes limitations of existing models focused only on simple graphs.

Abstract

Graph structures offer a versatile framework for representing diverse patterns in nature and complex systems, applicable across domains like molecular chemistry, social networks, and transportation systems. While diffusion models have excelled in generating various objects, generating graphs remains challenging. This thesis explores the potential of score-based generative models in generating such objects through a modelization as combinatorial complexes, which are powerful topological structures that encompass higher-order relationships. In this thesis, we propose a unified framework by employing stochastic differential equations. We not only generalize the generation of complex objects such as graphs and hypergraphs, but we also unify existing generative modelling approaches such as Score Matching with Langevin dynamics and Denoising Diffusion Probabilistic Models. This innovation overcomes limitations in existing frameworks that focus solely on graph generation, opening up new possibilities in generative AI. The experiment results showed that our framework could generate these complex objects, and could also compete against state-of-the-art approaches for mere graph and molecule generation tasks.