Stabilizing Filters for Graph Signal Generation
Abstract
Generating graph signals is an important task in machine learning, and incorporating the underlying graph into a generative model enables structurally-informed distribution modeling. However, said graph is often noisy or imperfectly estimated, and structural errors can affect the generative process and the resulting signal distribution. While recent work has established stability guarantees for graph neural ordinary differential equations, the robustness of graph-aware reference dynamics, as used in topological Schrödinger bridge models, remains largely unexplored. Here, we investigate the structural stability of continuous-time generative models whose dynamics combine a graph filter with a learned graph neural network. We first derive a Wasserstein stability bound that quantifies the effect of graph perturbations on the generated signal distributions. Informed by this analysis, we then propose a principled framework for designing graph filters that balance structural stability with induced signal smoothing. We characterize the optimal filter design and provide an explicit family of polynomial filters with provable stability guarantees. Experiments on synthetic graphs and real-world fMRI signals support our theoretical findings, illustrating the enhanced structural stability of the proposed filters while preserving generative performance.
Related
- Signal Processing over Product DAGs: Causal Shifts and Filters
- Stable Filters for Generative Modeling of Graph Signals
- Graph neural networks for sampling-invariant embeddings of organized signal sets
- hp-adaptive trees for graph signal approximation
- From topology learning to graph generation: A unifying perspective
BibTeX
@misc{schmidt2026stabilizing,
title = {{Stabilizing Filters for Graph Signal Generation}},
author = {Martin Schmidt and Gonzalo Mateos},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.28318},
eprint = {2610.28318},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.28318},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Stabilizing Filters for Graph Signal Generation
AU - Martin Schmidt
AU - Gonzalo Mateos
PY - 2026
DA - 2026/10/20
PB - arXiv
DO - 10.48550/arxiv.2610.28318
UR - https://arxiv.org/abs/2610.28318
AB - Generating graph signals is an important task in machine learning, and incorporating the underlying graph into a generative model enables structurally-informed distribution modeling. However, said graph is often noisy or imperfectly estimated, and structural errors can affect the generative process and the resulting signal distribution. While recent work has established stability guarantees for graph neural ordinary differential equations, the robustness of graph-aware reference dynamics, as used in topological Schrödinger bridge models, remains largely unexplored. Here, we investigate the structural stability of continuous-time generative models whose dynamics combine a graph filter with a learned graph neural network. We first derive a Wasserstein stability bound that quantifies the effect of graph perturbations on the generated signal distributions. Informed by this analysis, we then propose a principled framework for designing graph filters that balance structural stability with induced signal smoothing. We characterize the optimal filter design and provide an explicit family of polynomial filters with provable stability guarantees. Experiments on synthetic graphs and real-world fMRI signals support our theoretical findings, illustrating the enhanced structural stability of the proposed filters while preserving generative performance.
ER -