Meshless Latent Dynamics Models for Model Order Reduction of time-dependent Partial Differential Equations
Organizations: mathLab, Mathematics Area, SISSA, via Bonomea 265, I-34136 Trieste, Italy
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.
Figures & tables
| Causality | Time continuity | Multifidelity | Off-grid evaluation | # Params | |
|---|---|---|---|---|---|
| Neural operator | ✓ | ✓ | |||
| Autoregressive | ✓ | ✗ | |||
| Vanilla CNeF | ✗ | ✗ | ✓ | ✓ | |
| Mesh-based LDM | ✓ | ✓ | ✗ | ✗ | |
| Meshless LDM | ✓ | ✓ | ✓ | ✓ |
| Mesh-based | Meshless | ||||
|---|---|---|---|---|---|
| LD-FNN | LD-GCN | LDNet | FiLM-LDM | LDNet (0) | |
| Training time (s) | |||||
| / | |||||
| Mesh-based | Meshless | ||||
|---|---|---|---|---|---|
| LD-FNN | ( LDNet ) | ||||
| Training time (s) | |||||
| / | |||||
Appendix figures & tables2 assets
Supplementary material from the paper’s appendix.
Appendix
| Conditioning type | # Trainable parameters | # Operations |
|---|---|---|
| Weak | ||
| Strong | ||
| Pure |
| Architecture | Epochs | |||||
| LD-FNN (Glorot) | ✗ | 1500$+$1000 | ✗ | |||
| LD-FNN (He) | ✗ | 1500$+$1000 | ✗ | |||
| Burgers | LDNet | 1500$+$1000 | ||||
| LD-FNN | 25 | ✗ | 2000$+$1000 | ✗ | ||
| LD-GCN | 25 | ✗ | 2000$+$1000 | ✗ | ||
| LDNet | 25 | 2000$+$1000 |
Latent Dynamics Modeling
Latent dynamics modeling learns compact hidden-state representations and transition rules to predict, explain, and control evolving systems.
See all 274 papers