Abstract
Time-dependent partial differential equations (PDEs) are ubiquitous in science and engineering. Recently, mostly due to the high computational cost of traditional solution techniques, deep neural network based surrogates have gained increased interest. The practical utility of such neural PDE solvers relies on their ability to provide accurate, stable predictions over long time horizons, which is a notoriously hard problem. In this work, we present a large-scale analysis of common temporal rollout strategies, identifying the neglect of non-dominant spatial frequency information, often associated with high frequencies in PDE solutions, as the primary pitfall limiting stable, accurate rollout performance. Based on these insights, we draw inspiration from recent advances in diffusion models to introduce PDE-Refiner; a novel model class that enables more accurate modeling of all frequency components via a multistep refinement process. We validate PDE-Refiner on challenging benchmarks of complex fluid dynamics, demonstrating stable and accurate rollouts that consistently outperform state-of-the-art models, including neural, numerical, and hybrid neural-numerical architectures. We further demonstrate that PDE-Refiner greatly enhances data efficiency, since the denoising objective implicitly induces a novel form of spectral data augmentation. Finally, PDE-Refiner's connection to diffusion models enables an accurate and efficient assessment of the model's predictive uncertainty, allowing us to estimate when the surrogate becomes inaccurate.
| Originalsprache | Englisch |
|---|---|
| Titel | Conference Neural Information Processing Systems Foundation (NeurIPS 2023) |
| Seitenumfang | 1 |
| Publikationsstatus | Veröffentlicht - 2023 |
Wissenschaftszweige
- 305907 Medizinische Statistik
- 202017 Embedded Systems
- 202036 Sensorik
- 101004 Biomathematik
- 101014 Numerische Mathematik
- 101015 Operations Research
- 101016 Optimierung
- 101017 Spieltheorie
- 101018 Statistik
- 101019 Stochastik
- 101024 Wahrscheinlichkeitstheorie
- 101026 Zeitreihenanalyse
- 101027 Dynamische Systeme
- 101028 Mathematische Modellierung
- 101029 Mathematische Statistik
- 101031 Approximationstheorie
- 102 Informatik
- 102001 Artificial Intelligence
- 102003 Bildverarbeitung
- 102004 Bioinformatik
- 102013 Human-Computer Interaction
- 102018 Künstliche Neuronale Netze
- 102019 Machine Learning
- 102032 Computational Intelligence
- 102033 Data Mining
- 305901 Computerunterstützte Diagnose und Therapie
- 305905 Medizinische Informatik
- 202035 Robotik
- 202037 Signalverarbeitung
- 103029 Statistische Physik
- 106005 Bioinformatik
- 106007 Biostatistik
JKU-Schwerpunkte
- Digital Transformation
Dieses zitieren
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver