Generalized Spherical Neural Operators: Green's Function Formulation

2025/12/11 by Hao Tang, Hao Chen, Tang, Hao +3
Mathematics · Medicine · Physics and Astronomy · #Advanced Neuroimaging Techniques and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Model Reduction and Neural Networks #Tensor decomposition and applications

paper · pdf · doi:10.48550/arxiv.2512.10723

openalex publication_date 2025/12/11 · openalex created_date 2025/12/13 · openalex updated_date 2026/07/28

Abstract

Neural operators offer powerful approaches for solving parametric partial differential equations, but extending them to spherical domains remains challenging due to the need to preserve intrinsic geometry while avoiding distortions that break rotational consistency. Existing spherical operators rely on rotational equivariance but often lack the flexibility for real-world complexity. We propose a generalized operator-design framework based on the designable spherical Green's function and its harmonic expansion, establishing a solid operator-theoretic foundation for spherical learning. Based on this, we propose an absolute and relative position-dependent Green's function that enables flexible balance of equivariance and invariance for real-world modeling. The resulting operator, Green's-function Spherical Neural Operator (GSNO) with a novel spectral learning method, can adapt to non-equivariant systems while retaining spectral efficiency and grid invariance. To exploit GSNO, we develop SHNet, a hierarchical architecture that combines multi-scale spectral modeling with spherical up-down sampling, enhancing global feature representation. Evaluations on diffusion MRI, shallow water dynamics, and global weather forecasting, GSNO and SHNet consistently outperform state-of-the-art methods. The theoretical and experimental results position GSNO as a principled and generalized framework for spherical operator design and learning, bridging rigorous theory with real-world complexity. The code is available at: https://github.com/haot2025/GSNO.

Citations

• Sensitivity-Constrained Fourier Neural Operators for Forward and Inverse Problems in Parametric Differential Equations
• Equivariant Spherical CNNs for Accurate Fiber Orientation Distribution Estimation in Neonatal Diffusion MRI with Reduced Acquisition Time
• Enhancing Fourier Neural Operators with Local Spatial Features
• MscaleFNO: Multi-scale Fourier Neural Operator Learning for Oscillatory Function Spaces
• M²M: Learning controllable Multi of experts and multi-scale operators are the Partial Differential Equations need
• Improving Equivariant Model Training via Constraint Relaxation
• Huge Ensembles Part I: Design of Ensemble Weather Forecasts using Spherical Fourier Neural Operators
• Huge Ensembles Part II: Properties of a Huge Ensemble of Hindcasts Generated with Spherical Fourier Neural Operators
• Neural Operators with Localized Integral and Differential Kernels
• Operator Learning: Algorithms and Analysis
• Truly Scale-Equivariant Deep Nets with Fourier Layers
• Idealized Baroclinic Wave Test Results
• FAENet: Frame Averaging Equivariant GNN for Materials Modeling
• ClimaX: A foundation model for weather and climate
• Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions
• Equivariance and Invariance Inductive Bias for Learning from Insufficient Data
• Multiscale Neural Operator: Learning Fast and Grid-independent PDE Solvers
• Equivariant Graph Mechanics Networks with Constraints
• FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators
• Geometrically Equivariant Graph Neural Networks: A Survey
• Approximately Equivariant Networks for Imperfectly Symmetric Dynamics
• Residual Pathway Priors for Soft Equivariance Constraints
• Physics-Informed Neural Operator for Learning Partial Differential Equations
• Alias-Free Generative Adversarial Networks
• DISCO: accurate Discrete Scale Convolutions
• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• Fourier Neural Operator for Parametric Partial Differential Equations
• Multipole Graph Neural Operator for Parametric Partial Differential Equations
• Model Reduction and Neural Networks for Parametric PDEs
• A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects
• Neural Operator: Graph Kernel Network for Partial Differential Equations
• Spherical U-Net on Cortical Surfaces: Methods and Applications
• Spherical CNNs
• U-Net: Convolutional Networks for Biomedical Image Segmentation
• Scale-Invariant Convolutional Neural Networks
• Computing Fourier Transforms and Convolutions on the 2-Sphere
• Convolutional Neural Operators for robust and accurate learning of PDEs

Related

• Idealized Baroclinic Wave Test Results
• PDO-eS2CNNs: Partial Differential Operator Based Equivariant Spherical CNNs
• Efficient Generalized Spherical CNNs
• Spin-Weighted Spherical CNNs
• Scaling Spherical CNNs