Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Otness, Karl, Zanna, Laure, Bruna, Joan
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2303.17496
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915695303852032
author Otness, Karl
Zanna, Laure
Bruna, Joan
author_facet Otness, Karl
Zanna, Laure
Bruna, Joan
contents We propose a multiscale approach for predicting quantities in dynamical systems which is explicitly structured to extract information in both fine-to-coarse and coarse-to-fine directions. We envision this method being generally applicable to problems with significant self-similarity or in which the prediction task is challenging and where stability of a learned model's impact on the target dynamical system is important. We evaluate our approach on a climate subgrid parameterization task in which our multiscale networks correct chaotic underlying models to reflect the contributions of unresolved, fine-scale dynamics.
format Preprint
id arxiv_https___arxiv_org_abs_2303_17496
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Data-driven multiscale modeling for correcting dynamical systems
Otness, Karl
Zanna, Laure
Bruna, Joan
Atmospheric and Oceanic Physics
Machine Learning
We propose a multiscale approach for predicting quantities in dynamical systems which is explicitly structured to extract information in both fine-to-coarse and coarse-to-fine directions. We envision this method being generally applicable to problems with significant self-similarity or in which the prediction task is challenging and where stability of a learned model's impact on the target dynamical system is important. We evaluate our approach on a climate subgrid parameterization task in which our multiscale networks correct chaotic underlying models to reflect the contributions of unresolved, fine-scale dynamics.
title Data-driven multiscale modeling for correcting dynamical systems
topic Atmospheric and Oceanic Physics
Machine Learning
url https://arxiv.org/abs/2303.17496