DiGress: Discrete Denoising diffusion for graph generation

By Clément Vignac, Igor Krawczuk, Antoine Siraudin, Bohan Wang, Volkan Cevher, Pascal Frossard. International Conference on Learning Representations, 2022

670 citations recorded by the source corpus.

This work introduces DiGress, a discrete denoising diffusion model for generating graphs with categorical node and edge attributes. Our model utilizes a discrete diffusion process that progressively edits graphs with noise, through the process of adding or removing edges and changing the categories. A graph transformer network is trained to revert this process, simplifying the problem of distribution learning over graphs into a sequence of node and edge classification tasks. We further improve sample quality by introducing a Markovian noise model that preserves the marginal distribution of node and edge types during diffusion, and by incorporating auxiliary graph-theoretic features. A procedure for conditioning the generation on graph-level features is also proposed. DiGress achieves state-of-the-art performance on molecular and non-molecular datasets, with up to 3x validity improvement on a planar graph dataset. It is also the first model to scale to the large GuacaMol dataset containing 1.3M drug-like molecules without the use of molecule-specific representations.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Materials Chemistry

View Paper: https://www.semanticscholar.org/paper/c65c4b68ac153176354a4c33c37f0ba1d86772c0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Materials Chemistry

Citations in the corpus

Large Language Models on Graphs: A Comprehensive Survey (2023, neutral): “Graph Summarization, FAQ Generation, Query Suggestions — [126] Node Classification Given a graph G , predict the class of a node v .”
Simple and Effective Masked Diffusion Language Models (2024, background): “Diffusion models excel at producing realistic, high-quality images and have received significant attention as potential tools for generating discrete data such as text [1, 23, 25], biological sequences [2], and graphs [45, 48].”
Simplified and Generalized Masked Diffusion for Discrete Data (2024, positive): “…models have been extended to discrete state spaces [1, 14, 15] and have been successfully applied to applications ranging from graph generation [16], text-to-sound generation [17] or protein design [18], they remain not as widely used as their continuous counterparts as they are not…”
Equivariant 3D-conditional diffusion model for molecular linker design (2022, neutral): “More details on datasets, baselines and metrics can be found in Methods.”
A Survey on Generative Diffusion Models (2022, methodology): “In Section III-B3, we have previously mentioned the conventional methods for graph generation, which involve generating an adjacency matrix or node features through discrete diffusion [21], [22], [206].”