Neural partial differential equations for chaotic systems

Autor(es)

Gelbrecht, Maximilian ^{[1, 2, 3]} ; Boers, Niklas ^{[1, 4, 5, 2]} ; Kurths, Juergen ^{[1, 3, 6]}

Afiliação do(s) autor(es)

^[1] Potsdam Inst Climate Impact Res, Potsdam - Germany

^[2] Free Univ Berlin, Dept Math & Comp Sci, Berlin - Germany

^[3] Humboldt Univ, Phys Dept, Berlin - Germany

^[4] Univ Exeter, Global Syst Inst, Exeter, Devon - England

^[5] Univ Exeter, Dept Math, Exeter, Devon - England

^[6] Lobachevsky State Univ Nizhni Novgorod, Nizhnii Novgorod - Russia

• Tipo de documento: Artigo Científico
• Fonte: NEW JOURNAL OF PHYSICS; v. 23, n. 4 APR 2021.
• Citações Web of Science: 0

Resumo

Video Abstract Video Abstract: Neural partial differential equations for chaotic systems When predicting complex systems one typically relies on differential equation which can often be incomplete, missing unknown influences 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. (AU)

• Processo FAPESP: 15/50122-0 - Fenômenos dinâmicos em redes complexas: fundamentos e aplicações
• Beneficiário: Elbert Einstein Nehrer Macau
• Modalidade de apoio: Auxílio à Pesquisa - Temático