Effective models and predictability of chaotic multiscale systems via machine learning
| Autor: | Angelo Vulpiani, Francesco Borra, Massimo Cencini |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Computer science MathematicsofComputing_NUMERICALANALYSIS Degrees of freedom (statistics) Chaotic FOS: Physical sciences Chaotic dynamical systems machine learning multiscale systems Machine learning computer.software_genre 01 natural sciences Machine Learning (cs.LG) 010305 fluids & plasmas multiscale systems LYAPUNOV ANALYSIS TURBULENCE Adiabatic theorem 0103 physical sciences Limit (mathematics) Predictability 010306 general physics business.industry Reservoir computing Nonlinear Sciences - Chaotic Dynamics Nonlinear Sciences - Adaptation and Self-Organizing Systems machine learning Scale separation Imperfect Artificial intelligence Chaotic Dynamics (nlin.CD) business Adaptation and Self-Organizing Systems (nlin.AO) computer Chaotic dynamical systems |
| Zdroj: | Physical review. E (Online) 102 (2020): 052203-1–052203-11. doi:10.1103/PhysRevE.102.052203 info:cnr-pdr/source/autori:Borra F.; Vulpiani A.; Cencini M./titolo:Effective models and predictability of chaotic multiscale systems via machine learning/doi:10.1103%2FPhysRevE.102.052203/rivista:Physical review. E (Online)/anno:2020/pagina_da:052203-1/pagina_a:052203-11/intervallo_pagine:052203-1–052203-11/volume:102 |
| Popis: | We scrutinize the use of machine learning, based on reservoir computing, to build data-driven effective models of multiscale chaotic systems. We show that, for a wide scale separation, machine learning generates effective models akin to those obtained using multiscale asymptotic techniques and, remarkably, remains effective in predictability also when the scale separation is reduced. We also show that predictability can be improved by hybridizing the reservoir with an imperfect model. 12 pages with 10 figures. Accepted in Physical Review E |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|