Researchers
Choose France Chair in AI, Directrice de Recherche (Research Director), INRIA Paris; Professor, University of Colorado Boulder; Encadrante (PhD Supervisor), EDITE, Sorbonne University
Anastase Charantonis is an assistant professor in deep learning at ENSIIE, detached for research at INRIA for the year 2023-2024. He is affiliated to the Lamme and LOCEAN labs and collaborates with LiP6. His research is on the intersection of geosciences and learning approaches. Current themes include: imputation of sea-surface satellite data, seasonal to sub-seasonal weather forecasting, post processing probabilistic forecasts, domain adaptation and downscaling. Past themes include inversion of oceanic vertical distributions of physical properties from sea-surface observations, data assimilation and deep learning combination and investigating physical constraining of deep learning models.
Guillaume is a starting researcher (SRP) in the ARCHES team. Guillaume completed his PhD at FAIR Paris and Sorbonne Université, focusing on zero-shot adaptation of pretrained diffusion models for text-driven image editing. Guillaume is interested in AI-based probabilistic weather prediction, with the aim to reliably predict the electricity output of renewable energies a few hours to a couple of days in advance.
Research Engineer
Renu Singh
Students
Clément Dauvilliers is a PhD candidate at ARCHES and Sorbonne Université. His focus is on the potential uses of modern ML techniques to forecast extreme weather events, such as tropical storms, heavy precipitation or intense heat waves. He especially studies the downstream tasks related to the top weather prediction models (PanguWeather, GraphCast, FengWu, …).
David Landry is a PhD candidate at ARCHES after a few years in the industry. He is affiliated to Sorbonne Université. He is interested in the consequences of new artificial intelligence methods for weather and climate modelling.
Zourkalaini Boubakar, INRIA/Sorbonne PhD student
Graham Clyne, INRIA/Sorbonne PhD student
Graham is a PhD candidate at ARCHES and Sorbonne Université. He is interested in working with ML to understand impacts of land-use change on the climate.
Julie is a PhD candidate at EDF R&D and INRIA Lille. Her work focuses on optimizing deep neural networks (AutoDL) with applications to energy time series forecasting and spatio-temporal forecasting. She is interested in topics related to deep learning models designed for load consumption forecasting, wind generation forecasting and climate-related topics.
Interns
Maya Janvier, M2 internship, 2024
Alban Derepas, M2 internship (with EDF), 2024
Visitors
Nidhin Harilal, University of Colorado Boulder, Summer 2024
As a PhD candidate at the Alfred-Wegener Institute and University of Bremen, Yvonne works at the interface between oceanography and data science and is part of the MarDATA graduate school. She uses physical and biogeochemical measurement data to develop methods that advance our understanding of the oceans. She is especially interested in unsupervised learning, including missing value imputation strategies and clustering approaches as well as their validation.
Ayoub Ghriss, University of Colorado Boulder, Summer 2023
Kerri Lu, MIT, Summer 2023
Partnerships & Collaborations
Climate and Machine Learning Boulder (CLIMB)
EDF Research
