From Basis to Basis: Gaussian Particle Representation for Interpretable PDE Operators

Zhihao Li, Yu Feng, Zhilu Lai, Wei Wang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:66885-66902, 2026.

Abstract

Learning fluid PDE dynamics has increasingly benefited from neural operators and Transformer-based models, but their latent states often remain opaque, and sample-wise attention can be costly at high resolutions. We propose the Gaussian Particle Operator (GPO), a basis-to-basis neural operator that represents fields with a learned Gaussian basis. Each atom carries explicit geometric parameters, including centers, anisotropic scales, and weights, yielding a compact, mesh-agnostic, and directly visualizable intermediate representation. GPO operates in modal space: learned Gaussian modal windows perform Petrov–Galerkin measurements, and a PG Gaussian Attention module couples the resulting modes globally before scattering them back to the spatial domain. This design is resolution-agnostic, scales near-linearly with the number of samples for a fixed modal budget, and naturally applies to irregular geometries and 3D domains. We separately evaluate Gaussian-basis reconstruction and operator prediction. Across standard PDE benchmarks and real reanalysis datasets, GPO achieves competitive accuracy against neural-operator and Transformer baselines, ranking best on most tasks and close to the best on the remainder. The learned particles and modal couplings further provide representation-level interpretability, positioning GPO as a useful intermediate representation rather than a one-to-one decomposition of physical structures. The code is available at https://github.com/lizhihao2022/GPO.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-li26aq, title = {From Basis to Basis: {G}aussian Particle Representation for Interpretable {PDE} Operators}, author = {Li, Zhihao and Feng, Yu and Lai, Zhilu and Wang, Wei}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {66885--66902}, 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/li26aq/li26aq.pdf}, url = {https://proceedings.mlr.press/v306/li26aq.html}, abstract = {Learning fluid PDE dynamics has increasingly benefited from neural operators and Transformer-based models, but their latent states often remain opaque, and sample-wise attention can be costly at high resolutions. We propose the Gaussian Particle Operator (GPO), a basis-to-basis neural operator that represents fields with a learned Gaussian basis. Each atom carries explicit geometric parameters, including centers, anisotropic scales, and weights, yielding a compact, mesh-agnostic, and directly visualizable intermediate representation. GPO operates in modal space: learned Gaussian modal windows perform Petrov–Galerkin measurements, and a PG Gaussian Attention module couples the resulting modes globally before scattering them back to the spatial domain. This design is resolution-agnostic, scales near-linearly with the number of samples for a fixed modal budget, and naturally applies to irregular geometries and 3D domains. We separately evaluate Gaussian-basis reconstruction and operator prediction. Across standard PDE benchmarks and real reanalysis datasets, GPO achieves competitive accuracy against neural-operator and Transformer baselines, ranking best on most tasks and close to the best on the remainder. The learned particles and modal couplings further provide representation-level interpretability, positioning GPO as a useful intermediate representation rather than a one-to-one decomposition of physical structures. The code is available at https://github.com/lizhihao2022/GPO.} }
Endnote
%0 Conference Paper %T From Basis to Basis: Gaussian Particle Representation for Interpretable PDE Operators %A Zhihao Li %A Yu Feng %A Zhilu Lai %A Wei Wang %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-li26aq %I PMLR %P 66885--66902 %U https://proceedings.mlr.press/v306/li26aq.html %V 306 %X Learning fluid PDE dynamics has increasingly benefited from neural operators and Transformer-based models, but their latent states often remain opaque, and sample-wise attention can be costly at high resolutions. We propose the Gaussian Particle Operator (GPO), a basis-to-basis neural operator that represents fields with a learned Gaussian basis. Each atom carries explicit geometric parameters, including centers, anisotropic scales, and weights, yielding a compact, mesh-agnostic, and directly visualizable intermediate representation. GPO operates in modal space: learned Gaussian modal windows perform Petrov–Galerkin measurements, and a PG Gaussian Attention module couples the resulting modes globally before scattering them back to the spatial domain. This design is resolution-agnostic, scales near-linearly with the number of samples for a fixed modal budget, and naturally applies to irregular geometries and 3D domains. We separately evaluate Gaussian-basis reconstruction and operator prediction. Across standard PDE benchmarks and real reanalysis datasets, GPO achieves competitive accuracy against neural-operator and Transformer baselines, ranking best on most tasks and close to the best on the remainder. The learned particles and modal couplings further provide representation-level interpretability, positioning GPO as a useful intermediate representation rather than a one-to-one decomposition of physical structures. The code is available at https://github.com/lizhihao2022/GPO.
APA
Li, Z., Feng, Y., Lai, Z. & Wang, W.. (2026). From Basis to Basis: Gaussian Particle Representation for Interpretable PDE Operators. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:66885-66902 Available from https://proceedings.mlr.press/v306/li26aq.html.

Related Material