Neural partial differential equations for chaotic systems

Gelbrecht, Maximilian; Boers, Niklas; Kurths, Jürgen

doi:10.1088/1367-2630/abeb90

アイテム詳細

登録内容を編集ファイル形式で保存

一時保存へ追加

タグ情報を表示リリース履歴を表示詳細要約

公開

学術論文

Neural partial differential equations for chaotic systems

Authors

/persons/resource/gelbrecht

Gelbrecht, Maximilian
Potsdam Institute for Climate Impact Research;

/persons/resource/Niklas.Boers

Boers, Niklas
Potsdam Institute for Climate Impact Research;

/persons/resource/Juergen.Kurths

Kurths, Jürgen
Potsdam Institute for Climate Impact Research;

URL

There are no locators available

フルテキスト (公開)

25408oa.pdf
(出版社版), 6MB

付随資料 (公開)

There is no public supplementary material available

引用

Gelbrecht, M., Boers, N., & Kurths, J. (2021). Neural partial differential equations for chaotic systems. New Journal of Physics, 23:. doi:10.1088/1367-2630/abeb90.

引用: https://publications.pik-potsdam.de/pubman/item/item_25408

要旨

When predicting complex systems one typically relies on differential equation which can often be incomplete, missing unknown infl uences or higher order effects. By augmenting the equations with artificial neural networks we can compensate these deficiencies. We show that this can be used to predict paradigmatic, high-dimensional chaotic partial differential equations even when only short and incomplete datasets are available. The forecast horizon for these high dimensional systems is about an order of magnitude larger than the length of the training data.