Boosting performance in Machine Learning of Turbulent and Geophysical Flows via scale separation

Davide Faranda (1, 2) , M. Vrac (1) , Pascal Yiou (1) , F. M. E. Pons (1) , A Hamid (1) , G Carella (1) , C. G. Ngoungue Langue (1) , S Thao (1) , V. Gautard (3)

1 LSCE - Laboratoire des Sciences du Climat et de l'Environnement [Gif-sur-Yvette]
2 London Mathematical Laboratory
3 IRFU - Institut de Recherches sur les lois Fondamentales de l'Univers

Recent advances in statistical learning have opened the possibility to forecast the behavior of chaotic systems using recurrent neural networks. In this letter we investigate the applicability of this framework to geophysical flows, known to be intermittent and turbulent. We show that both turbulence and intermittency introduce severe limitations on the applicability of recurrent neural networks, both for short term forecasts as well as for the reconstruction of the underlying attractor. We suggest that possible strategies to overcome such limitations should be based on separating the smooth large-scale dynamics, from the intermittent/turbulent features. We test these ideas on global sea-level pressure data for the past 40 years, a proxy of the atmospheric circulation dynamics.

Domaines: Systèmes désordonnés et réseaux de neurones [cond-mat.dis-nn]; Dynamique Chaotique [nlin.CD]; Géophysique [physics.geo-ph]; Intelligence artificielle [cs.AI]

Davide Faranda, M. Vrac, Pascal Yiou, F. M. E. Pons, A Hamid, et al.. Boosting performance in Machine Learning of Turbulent and Geophysical Flows via scale separation. 2019. ⟨hal-03341712v1⟩