Topology-Preserving Neural Operator Learning via Hodge Decomposition

Dongzhe Zheng, Tao Zhong, Christine Allen-Blanchette
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:164029-164060, 2026.

Abstract

In this paper, we study solution operators of physical field equations on geometric meshes from a function-space perspective. We reveal that Hodge orthogonality fundamentally resolves spectral interference by isolating unlearnable topological degrees of freedom from learnable geometric dynamics, enabling an additive approximation confined to structure-preserving subspaces. Building on Hodge theory and operator splitting, we derive a principled operator-level decomposition. The result is a Hybrid Eulerian-Lagrangian architecture with an algebraic-level inductive bias we call Hodge Spectral Duality (HSD). In our framework, we use discrete differential forms to capture topology-dominated components and an orthogonal auxiliary ambient space to represent complex local dynamics. Our method achieves superior accuracy and efficiency on geometric graphs with enhanced fidelity to physical invariants.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-zheng26aj, title = {Topology-Preserving Neural Operator Learning via Hodge Decomposition}, author = {Zheng, Dongzhe and Zhong, Tao and Allen-Blanchette, Christine}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {164029--164060}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/zheng26aj/zheng26aj.pdf}, url = {https://proceedings.mlr.press/v306/zheng26aj.html}, abstract = {In this paper, we study solution operators of physical field equations on geometric meshes from a function-space perspective. We reveal that Hodge orthogonality fundamentally resolves spectral interference by isolating unlearnable topological degrees of freedom from learnable geometric dynamics, enabling an additive approximation confined to structure-preserving subspaces. Building on Hodge theory and operator splitting, we derive a principled operator-level decomposition. The result is a Hybrid Eulerian-Lagrangian architecture with an algebraic-level inductive bias we call Hodge Spectral Duality (HSD). In our framework, we use discrete differential forms to capture topology-dominated components and an orthogonal auxiliary ambient space to represent complex local dynamics. Our method achieves superior accuracy and efficiency on geometric graphs with enhanced fidelity to physical invariants.} }
Endnote
%0 Conference Paper %T Topology-Preserving Neural Operator Learning via Hodge Decomposition %A Dongzhe Zheng %A Tao Zhong %A Christine Allen-Blanchette %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-zheng26aj %I PMLR %P 164029--164060 %U https://proceedings.mlr.press/v306/zheng26aj.html %V 306 %X In this paper, we study solution operators of physical field equations on geometric meshes from a function-space perspective. We reveal that Hodge orthogonality fundamentally resolves spectral interference by isolating unlearnable topological degrees of freedom from learnable geometric dynamics, enabling an additive approximation confined to structure-preserving subspaces. Building on Hodge theory and operator splitting, we derive a principled operator-level decomposition. The result is a Hybrid Eulerian-Lagrangian architecture with an algebraic-level inductive bias we call Hodge Spectral Duality (HSD). In our framework, we use discrete differential forms to capture topology-dominated components and an orthogonal auxiliary ambient space to represent complex local dynamics. Our method achieves superior accuracy and efficiency on geometric graphs with enhanced fidelity to physical invariants.
APA
Zheng, D., Zhong, T. & Allen-Blanchette, C.. (2026). Topology-Preserving Neural Operator Learning via Hodge Decomposition. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:164029-164060 Available from https://proceedings.mlr.press/v306/zheng26aj.html.

Related Material