Graph Signal Generative Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Uslu, Yigit Berkay, Hadou, Samar, Rozada, Sergio, Bidokhti, Shirin Saeedi, Ribeiro, Alejandro
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916959928451072
author Uslu, Yigit Berkay
Hadou, Samar
Rozada, Sergio
Bidokhti, Shirin Saeedi
Ribeiro, Alejandro
author_facet Uslu, Yigit Berkay
Hadou, Samar
Rozada, Sergio
Bidokhti, Shirin Saeedi
Ribeiro, Alejandro
contents We introduce U-shaped encoder-decoder graph neural networks (U-GNNs) for stochastic graph signal generation using denoising diffusion processes. The architecture learns node features at different resolutions with skip connections between the encoder and decoder paths, analogous to the convolutional U-Net for image generation. The U-GNN is prominent for a pooling operation that leverages zero-padding and avoids arbitrary graph coarsening, with graph convolutions layered on top to capture local dependencies. This technique permits learning feature embeddings for sampled nodes at deeper levels of the architecture that remain convolutional with respect to the original graph. Applied to stock price prediction -- where deterministic forecasts struggle to capture uncertainties and tail events that are paramount -- we demonstrate the effectiveness of the diffusion model in probabilistic forecasting of stock prices.
format Preprint
id arxiv_https___arxiv_org_abs_2509_17250
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Graph Signal Generative Diffusion Models
Uslu, Yigit Berkay
Hadou, Samar
Rozada, Sergio
Bidokhti, Shirin Saeedi
Ribeiro, Alejandro
Machine Learning
Signal Processing
We introduce U-shaped encoder-decoder graph neural networks (U-GNNs) for stochastic graph signal generation using denoising diffusion processes. The architecture learns node features at different resolutions with skip connections between the encoder and decoder paths, analogous to the convolutional U-Net for image generation. The U-GNN is prominent for a pooling operation that leverages zero-padding and avoids arbitrary graph coarsening, with graph convolutions layered on top to capture local dependencies. This technique permits learning feature embeddings for sampled nodes at deeper levels of the architecture that remain convolutional with respect to the original graph. Applied to stock price prediction -- where deterministic forecasts struggle to capture uncertainties and tail events that are paramount -- we demonstrate the effectiveness of the diffusion model in probabilistic forecasting of stock prices.
title Graph Signal Generative Diffusion Models
topic Machine Learning
Signal Processing
url https://arxiv.org/abs/2509.17250