Stability and Discretization Error of State Space Model Neural Operators
Abstract
Neural operators have emerged as a powerful, discretization-invariant framework for solving partial differential equations (PDEs). Although established approaches like the Deep Operator Network (DeepONet) have successfully achieved universal approximation for operators, and architectures such as Fourier Neural Operators (FNOs) have shown algebraic convergence rates, a precise theoretical connection between the continuous theory and its discrete numerical implementation remains a challenge. Specifically, the relationship between the continuous formulation and the discrete numerical stability has yet to be fully explored. In this paper, we address this gap by establishing theoretical guarantees for the discretization error and stability of neural operator approximation schemes. We prove analytical bounds that link solution regularity to input discretization, providing a formal quantification of neural operator accuracy under real-world numerical constraints. We derive these bounds to the specific cases of State Space Model-based Neural Operators (SS-NOs) and FNOs, thus providing a new discretization error theorem for these models. Additionally, through an input-to-state stability (ISS) analysis, we formally assess the impact of discretization on the stability of SS-NOs results obtained in the continuous domain. Our empirical experiments on 1D and 2D benchmarks validate our theoretical bounds and show the robustness of SS-NOs under varying resolutions.
Related
- Symplectic-inspired neural operators for discretization-invariant light propagation with exact power conservation
- Fractional Laplace Neural Operators: Exact Architectures, an Expressivity Frontier at Criticality, and Certified Stability for Memory-Driven Network Dynamics
- Stochastic Gradient Leimkuhler-Matthews Integrator with Second Order Discretization Error
- Spatial Discretization Error for Quantum Chemistry Simulation
- Randomly shifted lattice rules for SDEs with constant diffusion: error bounds uniform in time discretization and dimension
BibTeX
@misc{bendahi2026stability,
title = {{Stability and Discretization Error of State Space Model Neural Operators}},
author = {Abderrahim Bendahi and Adrien Fradin and Johan Peralez and Julie Digne and Madiha Nadri},
howpublished = {HAL},
year = {2026},
month = may,
doi = {10.48550/arxiv.2605.18905},
eprint = {2605.18905},
archivePrefix = {arXiv},
url = {https://hal.science/hal-05829633v1},
note = {CC BY-SA 4.0},
}
RIS
TY - UNPB
TI - Stability and Discretization Error of State Space Model Neural Operators
AU - Abderrahim Bendahi
AU - Adrien Fradin
AU - Johan Peralez
AU - Julie Digne
AU - Madiha Nadri
PY - 2026
DA - 2026/05/17
PB - HAL
DO - 10.48550/arxiv.2605.18905
UR - https://hal.science/hal-05829633v1
AB - Neural operators have emerged as a powerful, discretization-invariant framework for solving partial differential equations (PDEs). Although established approaches like the Deep Operator Network (DeepONet) have successfully achieved universal approximation for operators, and architectures such as Fourier Neural Operators (FNOs) have shown algebraic convergence rates, a precise theoretical connection between the continuous theory and its discrete numerical implementation remains a challenge. Specifically, the relationship between the continuous formulation and the discrete numerical stability has yet to be fully explored. In this paper, we address this gap by establishing theoretical guarantees for the discretization error and stability of neural operator approximation schemes. We prove analytical bounds that link solution regularity to input discretization, providing a formal quantification of neural operator accuracy under real-world numerical constraints. We derive these bounds to the specific cases of State Space Model-based Neural Operators (SS-NOs) and FNOs, thus providing a new discretization error theorem for these models. Additionally, through an input-to-state stability (ISS) analysis, we formally assess the impact of discretization on the stability of SS-NOs results obtained in the continuous domain. Our empirical experiments on 1D and 2D benchmarks validate our theoretical bounds and show the robustness of SS-NOs under varying resolutions.
ER -