Can Deep Learning reveal Ito drift processes in upper ocean dynamics?

When:
31/12/2019 – 01/01/2020 all-day
2019-12-31T01:00:00+01:00
2020-01-01T01:00:00+01:00

Annonce en lien avec l’Action/le Réseau : aucun

Laboratoire/Entreprise : Lab-STICC/IMT Atlantique
Durée : 6 mois
Contact : ronan.fablet@imt-atlantique.fr
Date limite de publication : 31-12-2019

Contexte :
The understanding and evaluation of the impacts of fine-scale random processes onto larger-scale processes is in a key challenge in in physical oceanography. Ito-Wentzell formula provides the basic background to investigate these issues from a theoretical and computational point of view.

This internship is proposed in the framework of ANR Melody (Bridging geophysics and MachinE Learning for the modeling, simulation and reconstruction of Ocean DYnamics, PI: R. Fablet) and ERC STUOD (Stochastic Transport in Upper Ocean Dynamics, PI: B. Chapron).

Sujet :
The goal of this internship will be to investigate how neural ODE schemes could help revealing and understanding from data large-scale drift and diffusion processes caused by fine-scale processes in the upper ocean. For different case-studies, experiments will be performed on numerical simulations (e.g., toy models, reduced ocean models, HR ocean simulations). Experiments on real observation datasets would also be of interest in a second step.

Profil du candidat :
Msc./Eng. degree in Applied Math., Data Science and/or Physical Oceanography with a good background in applied statistics.

Formation et compétences requises :
Good scientific programming skills (eg, Python)
First experience and/or knowledge on deep learning models and frameworks would be a plus.

Adresse d’emploi :
Lab-STICC, IMT Atlantique, Technopôle Brest-Iroise
29238 Brest cedex

Document attaché : internship_proposal_ItoOcean2019.pdf