Model reduction of parametric ordinary differential equations via autoencoders: representation properties and convergence analysis
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915938329165824 |
|---|---|
| author | Ballini, Enrico Gambarini, Marco Fumagalli, Alessio Formaggia, Luca Scotti, Anna Zunino, Paolo |
| author_facet | Ballini, Enrico Gambarini, Marco Fumagalli, Alessio Formaggia, Luca Scotti, Anna Zunino, Paolo |
| contents | We propose a reduced-order modeling approach for nonlinear, parameter-dependent ordinary differential equations (ODE). Dimensionality reduction is achieved using nonlinear maps represented by autoencoders. The resulting low-dimensional ODE is then solved using standard integration in time schemes, and the high-dimensional solution is reconstructed from the low-dimensional one. We investigate the architecture of neural networks for constructing effective autoencoders that hold necessary properties to reconstruct the input manifold with exact representation capabilities. We study the convergence of the reduced-order model to the high-fidelity one. Numerical experiments show the robustness and accuracy of our approach in different scenarios, highlighting its effectiveness in highly complex and nonlinear settings without sacrificing accuracy. Moreover, we examine how the reduction influences the stability properties of the reconstructed high-dimensional solution. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_21280 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Model reduction of parametric ordinary differential equations via autoencoders: representation properties and convergence analysis Ballini, Enrico Gambarini, Marco Fumagalli, Alessio Formaggia, Luca Scotti, Anna Zunino, Paolo Numerical Analysis We propose a reduced-order modeling approach for nonlinear, parameter-dependent ordinary differential equations (ODE). Dimensionality reduction is achieved using nonlinear maps represented by autoencoders. The resulting low-dimensional ODE is then solved using standard integration in time schemes, and the high-dimensional solution is reconstructed from the low-dimensional one. We investigate the architecture of neural networks for constructing effective autoencoders that hold necessary properties to reconstruct the input manifold with exact representation capabilities. We study the convergence of the reduced-order model to the high-fidelity one. Numerical experiments show the robustness and accuracy of our approach in different scenarios, highlighting its effectiveness in highly complex and nonlinear settings without sacrificing accuracy. Moreover, we examine how the reduction influences the stability properties of the reconstructed high-dimensional solution. |
| title | Model reduction of parametric ordinary differential equations via autoencoders: representation properties and convergence analysis |
| topic | Numerical Analysis |
| url | https://arxiv.org/abs/2509.21280 |