Abstract:Score-based diffusion models in infinite-dimensional function spaces provide a mathematically principled framework for modelling function-valued data, offering key advantages such as resolution invariance and the ability to handle irregular discretisations. However, practical implementations have struggled to fully realise these benefits. Existing backbones like Fourier neural operators are often biased towards regular grids and fail to generalise to complex domain topologies. We propose a novel architecture for function-space diffusion models that represents generalised graph convolutional kernels as finite element functions, enabling the model to naturally handle unstructured meshes and complex geometries. We demonstrate the efficacy of our network architecture through a series of unconditional and conditional sampling experiments across diverse geometries, including non-convex and multiply-connected domains. Our results show that the proposed method maintains resolution invariance and achieves high fidelity in capturing functional distributions on non-trivial geometries.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.03497 [cs.LG] |
| (or arXiv:2605.03497v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.03497 arXiv-issued DOI via DataCite (pending registration) |
From: James Rowbottom [view email]
[v1]
Tue, 5 May 2026 08:33:52 UTC (6,549 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。