Julien Brajard
49
Documents
Presentation
My work is in the field of remote sensing, inverse modelling, machine learning and data assimilation.
The objective is to propose new methodologies in order to extract knowledge from data and physical systems,
more specifically in oceanography.
Those methodologies were apply to estimate and forecast key variables and their associated uncertainty in the ocean such as
phytoplankton and surface currents. The methodologies developed were
using remote sensing data (satellite sensors).
Key words : Machine learning, statistical methods, remotesensing, oceanography, computer science
Publications
|
Learning 4DVAR inversion directly from observationsWorkshop on Machine Learning and Data Assimilation for Dynamical Systems (MLDADS), International Conference on Computational Science (ICCS), Jul 2023, Prague, Czech Republic
Conference papers
hal-03863390v2
|
|
Variational assimilation of Geophysical images leveraging deep learning toolsORASIS 2021, Centre National de la Recherche Scientifique [CNRS], Sep 2021, Saint Ferréol, France
Conference papers
hal-03339674v1
|
|
Completing physics-based models by learning hidden dynamics through data assimilationNeurIPS 2020, workshop AI4Earth, Dec 2020, Vancouver (virtual), Canada
Conference papers
hal-03004938v1
|
Mesoscale Activity in the Eastern Mediterranean: Blending Altimetry with in situ observationsEGU General assembly 2020, May 2020, virtual, Zimbabwe. ⟨10.5194/egusphere-egu2020-20407⟩
Conference papers
hal-03082863v1
|
|
Can a neural network learn a numerical model ?EGU General Assembly, Apr 2018, Vienne, Austria
Conference papers
hal-02067473v1
|
|
Improving CADNA Performance on GPUs2018 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), May 2018, Vancouver, Canada. pp.1016-1025, ⟨10.1109/IPDPSW.2018.00156⟩
Conference papers
hal-01858537v1
|
|
Predicting Ocean Dynamics through Machine Learning: Application on Sea-Surface Suspended Particulate MaterAmerican Meteorological Society Annual Meeting 2017, American Meteorological Society, Jan 2017, Seattle, United States
Conference papers
hal-01477182v1
|
|
Ensemble variational data assimilation with a shallow-water model: preliminary resultsEGU General Assembly, Apr 2016, Vienne, Austria. pp.14836
Conference papers
hal-01507519v1
|
|
Determination of phytoplankton groups from space: application to senegalo-mauritanean upwellingEGU General Assembly, 2016, Vienne, Austria. pp.15098
Conference papers
hal-01507521v1
|
|
|
Estimation of Round-off Errors in OpenMP CodesIWOMP 2016 - 12th International Workshop on OpenMP, Riken AICS, Oct 2016, Nara, Japan. pp.3-16, ⟨10.1007/978-3-319-45550-1_1⟩
Conference papers
hal-01380131v1
|
|
Determination of phytoplankton groups from ocean color spectral measurements in the Senegalo-Mauritanian upwellingOcean Optics, The Oceanography Society (TOS), Oct 2014, Portland, Maine, United States
Conference papers
hal-01191676v1
|
Towards high performance stochastic arithmetic16th international symposium on Scientific Computing, Computer Arithmetic and Validated Numerics (SCAN 2014), Sep 2014, Würzburg, Germany. pp.47-48
Conference papers
hal-01217244v1
|
|
Numerical Validation of Data Assimilation Codes Generated by the YAO SoftwareSIAM Annual Meeting, Jul 2013, San Diego, CA, United States
Conference papers
hal-01216453v1
|
|
Neuro-Variational Inversion Applied to Ocean Color Remote SensingAmerican Meteorological Society Annual Meeting, 2012, La Nouvelle Orléans, United States
Conference papers
hal-00750316v1
|
|
Inverse method for the retrieval of ocean vertical profiles using self-organizing maps and hidden Markov models - Application on Ocean Colour Satellite Image InversionNCTA 2011 : NCTA- International Conference on Neural Computation Theory and Applications, 2011, Paris, France
Conference papers
hal-00750300v1
|
|
Retrieving aerosol characteristics from satellite ocean color multi-spectral sensors using a neural-variational methodNCTA 2011 : NCTA- International Conference on Neural Computation Theory and Applications, 2011, Paris, France
Conference papers
hal-00750307v1
|
|
Inversion of sea surface data to retrieve the vertical distribution of phytoplankton using a statistical methodologyCongrès European Geophysical Union, 2011, Vienne, Austria
Conference papers
hal-00750248v1
|
|
Variational data assimilation in NEMO using YAOCongrès European Geophysical Union, 2011, Vienne, Austria
Conference papers
hal-00750253v1
|
|
Atmospheric correction and chlorophyll-a restitution in presence of Saharan dustsCongrès European Geophysical Union, 2011, Vienne, Austria
Conference papers
hal-00750241v1
|
|
Modeling Surface Currents in the Eastern Levantine MediterraneanEGU General Assembly, Apr 2015, Vienna, Austria. 17, pp.8888
Conference poster
hal-01191755v1
|
|
Spatial Distribution of Aerosol Optical Thickness Retrieved from SeaWiFS Images by a Neural Network Inversion over the West African CoastAerosols-Science and Case Studies, InTech, 2016, 978-953-51-2843-4. ⟨10.5772/65874⟩
Book sections
hal-01507520v1
|
Learning Dynamical Systems from Partial Observations2019
Preprints, Working Papers, ...
hal-02418376v1
|
|
|
Automatic generation of parallel and coherent code using the YAO variational data assimilation framework2016
Preprints, Working Papers, ...
hal-00783328v2
|