Meshless Latent Dynamics Models for Model Order Reduction of time-dependent Partial Differential Equations
Abstract
Latent dynamics learning has emerged as a prominent approach for Model Order Reduction (MOR) of time-dependent, parameterized Partial Differential Equations (PDEs). Traditional autoencoder-based approaches are intrinsically tied to fixed spatial discretizations during both training and inference, motivating the development of meshless methodologies. In this context, conditioning provides a natural mechanism for representing temporal evolution through spatially continuous functions, enabling mesh-agnostic Latent Dynamics Models (LDMs). In this work, we introduce a unified mathematical framework for mesh-agnostic LDMs based on conditioning in Neural Fields (NeFs). Leveraging a taxonomy of conditioning mechanisms (weak, strong, and pure), we formulate meshless LDMs as NeFs conditioned on a latent state. We establish a Universal Approximation Theorem (UAT) for this class of models and analyze their stability by deriving Lipschitz bounds for different conditioning mechanisms. Our analysis reveals an expressivity--stability trade-off, providing insight into why more expressive conditioning, despite its higher representational capacity, can potentially degrade performance during temporal extrapolation. Finally, we provide a comprehensive comparison of meshless and mesh-based architectures in the continuous limit of the spatial discretization, showing that meshless models guarantee grid-independent Lipschitz bounds, unlike mesh-based counterparts. Our experiments across Burgers' equation, Navier--Stokes flow past a cylinder, and linear advection show that meshless LDMs can achieve predictive accuracy comparable to or better than their mesh-based counterparts while using only 2% of the trainable parameters and exhibiting nearly constant error across spatial discretizations.
Related
- Spectral Chebyshev wavelet method for state-dependent fractional integro-differential equations in adaptive tumor growth modeling
- Derivative-Informed Dimension Reduction of Stochastic Differential Equations: A Path-Space Sensitivity Approach
- Convergence Analysis of Variable-Separation-Based Model Reduction for Linear Parametric Partial Differential Equations
- Path-dependent rough differential equations in the full step-two range
- Bounds-Constrained Finite Element Approximation of Time-Dependent Partial Differential Equations
BibTeX
@misc{tomada2026meshless,
title = {{Meshless Latent Dynamics Models for Model Order Reduction of time-dependent Partial Differential Equations}},
author = {Lorenzo Tomada and Federico Pichi and Gianluigi Rozza},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.41043},
eprint = {2610.41043},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.41043},
}
RIS
TY - UNPB
TI - Meshless Latent Dynamics Models for Model Order Reduction of time-dependent Partial Differential Equations
AU - Lorenzo Tomada
AU - Federico Pichi
AU - Gianluigi Rozza
PY - 2026
DA - 2026/10/29
PB - arXiv
DO - 10.48550/arxiv.2610.41043
UR - https://arxiv.org/abs/2610.41043
AB - Latent dynamics learning has emerged as a prominent approach for Model Order Reduction (MOR) of time-dependent, parameterized Partial Differential Equations (PDEs). Traditional autoencoder-based approaches are intrinsically tied to fixed spatial discretizations during both training and inference, motivating the development of meshless methodologies. In this context, conditioning provides a natural mechanism for representing temporal evolution through spatially continuous functions, enabling mesh-agnostic Latent Dynamics Models (LDMs). In this work, we introduce a unified mathematical framework for mesh-agnostic LDMs based on conditioning in Neural Fields (NeFs). Leveraging a taxonomy of conditioning mechanisms (weak, strong, and pure), we formulate meshless LDMs as NeFs conditioned on a latent state. We establish a Universal Approximation Theorem (UAT) for this class of models and analyze their stability by deriving Lipschitz bounds for different conditioning mechanisms. Our analysis reveals an expressivity--stability trade-off, providing insight into why more expressive conditioning, despite its higher representational capacity, can potentially degrade performance during temporal extrapolation. Finally, we provide a comprehensive comparison of meshless and mesh-based architectures in the continuous limit of the spatial discretization, showing that meshless models guarantee grid-independent Lipschitz bounds, unlike mesh-based counterparts. Our experiments across Burgers' equation, Navier--Stokes flow past a cylinder, and linear advection show that meshless LDMs can achieve predictive accuracy comparable to or better than their mesh-based counterparts while using only 2% of the trainable parameters and exhibiting nearly constant error across spatial discretizations.
ER -