Efficient Dilated Squeeze and Excitation Neural Operator for Differential Equations
Preprint
Abstract
Fast and accurate surrogates for physics-driven partial differential equations (PDEs) are essential in fields such as aerodynamics, porous media design, and flow control. However, many transformer-based models and existing neural operators remain parameter-heavy, resulting in costly training and sluggish deployment. We propose D-SENO (Dilated Squeeze-Excitation Neural Operator), a lightweight operator learning framework for efficiently solving a wide range of PDEs, including airfoil potential flow, Darcy flow in porous media, pipe Poiseuille flow, and incompressible Navier Stokes vortical fields. D-SENO combines dilated convolution (DC) blocks with squeeze-and-excitation (SE) modules to jointly capture wide receptive fields and dynamics alongside channel-wise attention, enabling both accurate and efficient PDE inference. Carefully chosen dilation rates allow the receptive field to focus on critical regions, effectively modeling long-range physical dependencies. Meanwhile, the SE modules adaptively recalibrate feature channels to emphasize dynamically relevant scales. Our model achieves training speed of up to approximately $20\times$ faster than standard transformer-based models and neural operators, while also surpassing (or matching) them in accuracy across multiple PDE benchmarks. Ablation studies show that removing the SE modules leads to a slight drop in performance.
Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Neural operators struggle to learn complex PDEs in pedestrian mobility: Hughes model case studyArtificial Intelligence for Transportation, 2025
- Dilated convolution neural operator for multiscale partial differential equationsJournal of Computational and Applied Mathematics, 2024
- Scalable Transformer for PDE Surrogate Modeling2023
- Wavelet Neural Operator for solving parametric partial differential equations in computational mechanics problemsComputer Methods in Applied Mechanics and Engineering, 2022
- U-FNO—An enhanced Fourier neural operator-based deep-learning model for multiphase flowAdvances in Water Resources, 2022
- Learning nonlinear operators via DeepONet based on the universal approximation theorem of operatorsNature Machine Intelligence, 2021
- Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equationsJournal of Computational Physics, 2019