CompNO: A Novel Foundation Model approach for solving Partial Differential Equations
Authors:
Preprint
Abstract
Partial differential equations (PDEs) govern a wide range of physical phenomena, but their numerical solution remains computationally demanding, especially when repeated simulations are required across many parameter settings. Recent Scientific Foundation Models (SFMs) aim to alleviate this cost by learning universal surrogates from large collections of simulated systems, yet they typically rely on monolithic architectures with limited interpretability and high pretraining expense. In this work we introduce Compositional Neural Operators (CompNO), a compositional neural operator framework for parametric PDEs. Instead of pretraining a single large model on heterogeneous data, CompNO first learns a library of Foundation Blocks, where each block is a parametric Fourier neural operator specialized to a fundamental differential operator (e.g. convection, diffusion, nonlinear convection). These blocks are then assembled, via lightweight Adaptation Blocks, into task-specific solvers that approximate the temporal evolution operator for target PDEs. A dedicated boundary-condition operator further enforces Dirichlet constraints exactly at inference time. We validate CompNO on one-dimensional convection, diffusion, convection--diffusion and Burgers' equations from the PDEBench suite. The proposed framework achieves lower relative L2 error than strong baselines (PFNO, PDEFormer and in-context learning based models) on linear parametric systems, while remaining competitive on nonlinear Burgers' flows. The model maintains exact boundary satisfaction with zero loss at domain boundaries, and exhibits robust generalization across a broad range of Peclet and Reynolds numbers. These results demonstrate that compositional neural operators provide a scalable and physically interpretable pathway towards foundation models for PDEs.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Fine-tune language models as multi-modal differential equation solversNeural Networks, 2025
- Kolmogorov–Arnold-Informed neural network: A physics-informed deep learning framework for solving forward and inverse problems based on Kolmogorov–Arnold NetworksComputer Methods in Applied Mechanics and Engineering, 2024
- PROSE: Predicting Multiple Operators and Symbolic Expressions using multimodal transformersNeural Networks, 2024
- Parametric learning of time-advancement operators for unstable flame evolutionPhysics of Fluids, 2024
- In-context operator learning with data prompts for differential equation problemsProceedings of the National Academy of Sciences, 2023
- A Deep Fourier Residual method for solving PDEs using Neural NetworksComputer Methods in Applied Mechanics and Engineering, 2022
- Learning by neural networks under physical constraints for simulation in fluid mechanicsComputers & Fluids, 2022
- Prediction of aerodynamic flow fields using convolutional neural networksComputational Mechanics, 2019
- Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equationsJournal of Computational Physics, 2019
- Bayesian deep convolutional encoder–decoder networks for surrogate modeling and uncertainty quantificationJournal of Computational Physics, 2018