Organizations: Financial University under the Government of the Russian Federation, Leningradsky Prospect, 49, Moscow, 109456, Russia. · The Scientific Research Institute of Goznak, Mytnaya Str. 17, Moscow, 115162, Russia. · MIREA—Russian Technological University 78 Vernadsky Avenue, Moscow, 119454, Russia. · Central Economics and Mathematics Institute of the Russian Academy of Sciences , Nakhimovsky Prospect, 47, Moscow, 117418, Russia.
The scientific computation methods development in conjunction with artificial intelligence technologies remains a hot research topic. Finding a balance between lightweight and accurate computations is a solid foundation for this direction. The study presents a neural operator based on the dynamic mode decomposition algorithm (DMD), mapping functional spaces, which combines DMD and deep learning (DL) for spatiotemporal processes efficient modeling. Solving PDEs for various initial and boundary conditions requires significant computational resources. The method suggested automatically extracts key modes and system dynamics using them to construct predictions, reducing computational costs compared to traditional numerical methods. The approach has demonstrated its efficiency through comparative analysis of performance with closest analogues DeepONet and FNO in the heat equation, Laplaces equation, and Burgers equation solutions approximation, where it achieves high reconstruction accuracy.
Figures & tables
Figure 1: Conceptual neural network architecture
Figure 2: Example implementation of neural network architecture
Parameter
Value
Number of samples
1000
Grid size
10 × 10
Number of iterations
50
Boundary value range
U∼(−10,10)
Fixed corner values
0 (Dirichlet condition)
Table 1: Data generation parameters for Laplace equation
Figure 3: Experimental results for the Laplace equation under various boundary conditions
Figure 4: Training and test loss functions for the Laplace equation
Training
Test
Epoch
Loss
Epoch
Loss
0
12.103361
0
10.639065
10
3.567372
10
3.901556
20
1.646200
20
3.254108
30
1.479754
30
3.473141
40
1.407965
40
1.460085
Table 2: Training and test losses over epochs for Laplace equation
Parameter
Value
Number of samples
1000
Spatial grid size
10 × 10
Number of time steps
50
Thermal diffusivity α
0.5
Random initial condition
Corners: U∼(−25,25)
Fixed random seed
Yes (for reproducibility)
Table 3: Data generation parameters
Figure 5: Experimental results for heat equation
Figure 6: Error functions for heat equation
Training
Test
Epoch
Loss
Epoch
Loss
0
83.191646
0
73.319000
10
2.536713
10
1.407479
20
1.244828
20
1.612512
30
1.072387
30
1.256115
40
0.866407
40
0.229214
Table 4: Heat Equation Training and Test Losses
Parameter
Value
Number of samples
1000
Grid size
10 × 10
Number of time steps
50
Initial condition range
U,V∼(−25,25)
Viscosity coefficient ν
U(0.01,0.1)
Time step Δt
0.0001
Table 5: Data generation parameters for Burgers’ equation
Figure 7: Experimental results for Burgers’ equation
Figure 8: Error functions for Burgers’ equation
Training
Test
Epoch
Loss
Epoch
Loss
0
78.371226
0
35.133656
10
0.300840
10
0.173647
20
0.250105
20
0.563705
30
0.089552
30
0.373777
40
0.066569
40
0.024333
Table 6: Burgers’ Equation Training and Test Losses
Neural operators have emerged as powerful surrogates for dynamical systems due to their grid-invariant properties and computational efficiency. However, Fourier-based variants inherently truncate high-frequency components in spectral space, resulting in the loss of small-scale structures and degraded prediction quality at high resolutions when trained on low-resolution data. While diffusion-based enhancement methods can recover multi-scale features, they introduce substantial inference overhead that undermines the efficiency advantage of neural operators. In this work, we introduce MeanFlow-Enhanced Neural Operators (MENO), a novel framework that achieves accurate all-scale predictions with minimal inference cost. By leveraging the improved MeanFlow method, MENO restores both small-scale details and large-scale dynamics with superior physical fidelity and statistical accuracy. We evaluate MENO on three challenging dynamical systems, including phase-field dynamics, 2D Kolmogorov flow, and active matter dynamics, at resolutions up to 256×256. Across all benchmarks, MENO improves the power spectrum density accuracy by up to a factor of 2 compared to baseline neural operators while achieving up to 14× faster inference than the state-of-the-art Denoising Diffusion Implicit Model (DDIM)-enhanced counterparts, effectively bridging the gap between accuracy and efficiency. The flexibility and efficiency of MENO position it as an efficient surrogate model for scientific machine learning applications where both statistical integrity and computational efficiency are paramount.
Tianyue Yang, Xiao Xue
Centre for Computational Science, University College London, London, United Kingdom
Partial differential equations (PDEs) govern complex systems, yet neural operators often struggle to efficiently capture the long-range, nonlocal interactions inherent in their solution maps. We introduce Spectral Filtering Operator (SFO), a neural operator that parameterizes integral kernels using the Universal Spectral Basis (USB), a fixed, global orthonormal basis derived from the eigenmodes of the Hilbert matrix in spectral filtering theory. Motivated by our theoretical finding that the discrete Green's functions of shift-invariant PDE discretizations exhibit spatial Linear Dynamical System (LDS) structure, we prove that these kernels admit compact approximations in the USB. By learning only the spectral coefficients of rapidly decaying eigenvalues, SFO achieves a highly efficient representation. Across six benchmarks, including reaction-diffusion, fluid dynamics, and 3D electromagnetics, SFO achieves state-of-the-art accuracy, reducing error by up to 40% relative to strong baselines while using substantially fewer parameters.
Noam Koren, Rafael Moschopoulos, Kira Radinsky +1
Department of Computer Science Technion - Israel Institute of Technology · Department of Computer Science Princeton University
Neural operators have emerged as promising surrogate models for solving partial differential equations (PDEs), but struggle to generalise beyond training distributions and are often constrained to a fixed temporal discretisation. This work introduces a physics-informed training framework that addresses these limitations by decomposing PDEs using operator splitting methods, training separate neural operators to learn individual non-linear physical operators while approximating linear operators with fixed finite-difference convolutions. This modular mixture-of-experts architecture enables generalisation to novel physical regimes by explicitly encoding the underlying operator structure. We formulate the modelling task as a neural ordinary differential equation (ODE) where these learned operators constitute the right-hand side, enabling continuous-in-time predictions through standard ODE solvers and implicitly enforcing PDE constraints. Demonstrated on incompressible and compressible Navier--Stokes equations, our approach achieves better convergence and superior performance when generalising to unseen physics. The method remains parameter-efficient, enabling temporal extrapolation beyond training horizons, and provides interpretable components whose behaviour can be verified against known physics.
Vignesh Gopakumar, Ander Gray, Dan Giles +5
UCL Centre for Artificial Intelligence · UK Atomic Energy Authority · LIX, CNRS, ´Ecole Polytechnique +2