Extreme events prediction from nonlocal ...
Type de document :
Compte-rendu et recension critique d'ouvrage
Titre :
Extreme events prediction from nonlocal partial information in a spatiotemporally chaotic microcavity laser
Auteur(s) :
Pammi, V. A. [Auteur]
Centre de Nanosciences et de Nanotechnologies [C2N]
Clerc, M. G. [Auteur]
Universidad de Santiago de Chile [Santiago] [USACH]
Coulibaly, S. [Auteur]
Laboratoire de Physique des Lasers, Atomes et Molécules - UMR 8523 [PhLAM]
Barbay, Sylvain [Auteur]
Centre de Nanosciences et de Nanotechnologies [C2N]
Centre de Nanosciences et de Nanotechnologies [C2N]
Clerc, M. G. [Auteur]
Universidad de Santiago de Chile [Santiago] [USACH]
Coulibaly, S. [Auteur]
Laboratoire de Physique des Lasers, Atomes et Molécules - UMR 8523 [PhLAM]
Barbay, Sylvain [Auteur]
Centre de Nanosciences et de Nanotechnologies [C2N]
Titre de la revue :
Physical Review Letters
Pagination :
223801
Éditeur :
American Physical Society
Date de publication :
2023-03-03
ISSN :
0031-9007
Discipline(s) HAL :
Science non linéaire [physics]/Dynamique Chaotique [nlin.CD]
Résumé en anglais : [en]
The forecasting of high-dimensional, spatiotemporal nonlinear systems has made tremendous progress with the advent of model-free machine learning techniques. However, in real systems it is not always possible to have all ...
Lire la suite >The forecasting of high-dimensional, spatiotemporal nonlinear systems has made tremendous progress with the advent of model-free machine learning techniques. However, in real systems it is not always possible to have all the information needed; only partial information is available for learning and forecasting. This can be due to insufficient temporal or spatial samplings, to inaccessible variables or to noisy training data. Here, we show that it is nevertheless possible to forecast extreme events occurrence in incomplete experimental recordings from a spatiotemporally chaotic microcavity laser using reservoir computing. Selecting regions of maximum transfer entropy, we show that it is possible to get higher forecasting accuracy using nonlocal data vs local data thus allowing greater warning times, at least twice the time horizon predicted from the nonlinear local Lyapunov exponent.Lire moins >
Lire la suite >The forecasting of high-dimensional, spatiotemporal nonlinear systems has made tremendous progress with the advent of model-free machine learning techniques. However, in real systems it is not always possible to have all the information needed; only partial information is available for learning and forecasting. This can be due to insufficient temporal or spatial samplings, to inaccessible variables or to noisy training data. Here, we show that it is nevertheless possible to forecast extreme events occurrence in incomplete experimental recordings from a spatiotemporally chaotic microcavity laser using reservoir computing. Selecting regions of maximum transfer entropy, we show that it is possible to get higher forecasting accuracy using nonlocal data vs local data thus allowing greater warning times, at least twice the time horizon predicted from the nonlinear local Lyapunov exponent.Lire moins >
Langue :
Anglais
Vulgarisation :
Non
Source :
Fichiers
- 2303.03097
- Accès libre
- Accéder au document
- document
- Accès libre
- Accéder au document
- 2303.03097.pdf
- Accès libre
- Accéder au document