Exploring Molecule Generation Using Latent Space Graph Diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pombala, Prashanth, Grossmann, Gerrit, Wolf, Verena
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910775665229824
author Pombala, Prashanth
Grossmann, Gerrit
Wolf, Verena
author_facet Pombala, Prashanth
Grossmann, Gerrit
Wolf, Verena
contents Generating molecular graphs is a challenging task due to their discrete nature and the competitive objectives involved. Diffusion models have emerged as SOTA approaches in data generation across various modalities. For molecular graphs, graph neural networks (GNNs) as a diffusion backbone have achieved impressive results. Latent space diffusion, where diffusion occurs in a low-dimensional space via an autoencoder, has demonstrated computational efficiency. However, the literature on latent space diffusion for molecular graphs is scarce, and no commonly accepted best practices exist. In this work, we explore different approaches and hyperparameters, contrasting generative flow models (denoising diffusion, flow matching, heat dissipation) and architectures (GNNs and E(3)-equivariant GNNs). Our experiments reveal a high sensitivity to the choice of approach and design decisions. Code is made available at github.com/Prashanth-Pombala/Molecule-Generation-using-Latent-Space-Graph-Diffusion.
format Preprint
id arxiv_https___arxiv_org_abs_2501_03696
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exploring Molecule Generation Using Latent Space Graph Diffusion
Pombala, Prashanth
Grossmann, Gerrit
Wolf, Verena
Machine Learning
Artificial Intelligence
Generating molecular graphs is a challenging task due to their discrete nature and the competitive objectives involved. Diffusion models have emerged as SOTA approaches in data generation across various modalities. For molecular graphs, graph neural networks (GNNs) as a diffusion backbone have achieved impressive results. Latent space diffusion, where diffusion occurs in a low-dimensional space via an autoencoder, has demonstrated computational efficiency. However, the literature on latent space diffusion for molecular graphs is scarce, and no commonly accepted best practices exist. In this work, we explore different approaches and hyperparameters, contrasting generative flow models (denoising diffusion, flow matching, heat dissipation) and architectures (GNNs and E(3)-equivariant GNNs). Our experiments reveal a high sensitivity to the choice of approach and design decisions. Code is made available at github.com/Prashanth-Pombala/Molecule-Generation-using-Latent-Space-Graph-Diffusion.
title Exploring Molecule Generation Using Latent Space Graph Diffusion
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2501.03696