Thursday, September 4, 2025

Time Event (+)
08:30 - 09:00 Welcome coffee  
09:00 - 09:10 Introductory speech (RESSTE / Geolearning chair)  
09:10 - 09:50 Plenary Talk (+)  
09:10 - 09:50 › Arctic Ocean data assimilation assisted by machine learning - Laurent Bertino, Nansen Environmental and Remote Sensing Center [Bergen]  
09:50 - 10:50 Machine learning (+)  
09:50 - 10:10 › Benchmarking probabilistic spatial machine learning models with complex sample distributions - Jérémy Rohmer, BRGM  
10:10 - 10:30 › Model selection for spatial regression with cross-validation adapted to dependent data. - Valentina Bastidas Schade, Université Paris Nanterre - Modal'X  
10:30 - 10:50 › Bayesian prediction of non-stationary spatial data: when diffusion generative models meet Gaussian random fields - Mike Pereira, Mines Paris - PSL (École nationale supérieure des mines de Paris)  
10:50 - 11:10 Coffee break  
11:10 - 11:50 Plenary Talk (+)  
11:10 - 11:50 › Regular Variation in Hilbert Spaces and Principal Component Analysis for Functional Extremes - Anne Sabourin, Mathématiques Appliquées Paris 5  
11:50 - 12:30 Climate and extremes (+)  
11:50 - 12:10 › Analyzing the dynamics of extreme events with marked point processes. - Antoine Heranval, Biostatistique et Processus Spatiaux  
12:10 - 12:30 › A complete characterization of indicator variograms and madograms - Xavier Emery, University of Chile  
12:30 - 14:00 Lunch  
14:00 - 14:40 Plenary Talk (+)  
14:00 - 14:40 › Hierarchical transfer learning with applications for electricity load forecasting - Yannig Goude, Laboratoire de Mathématiques d'Orsay  
14:40 - 15:40 Spatial (+)  
14:40 - 15:00 › Cartographie des expositions environnementales en France : krigeage des données de qualité des sols - Mamadou Bailo BALDE, INERIS  
15:00 - 15:20 › Efficient Bayesian spatially varying coefficients modeling for censored and clustered data using the vecchia approximation - Yacine IDIR, Ecoles des mines de Paris  
15:20 - 15:40 › Spline Interpolation on Riemaniann Manifolds - Charlie SIRE, Mines Paris, PSL University, Centre for geosciences and geoengineering  
15:40 - 16:00 Coffee break  
16:00 - 16:40 Plenary Talk (+)  
16:00 - 16:40 › Learning data assimilation from artificial intelligence - Marc Bocquet, Centre d'Enseignement et de Recherche en Environnement Atmosphérique  
16:40 - 16:50 Poster Pitch Session  
16:50 - 18:00 Poster (+)  
16:50 - 18:00 › SPDE kriging for alluvial thickness mapping : application at a regional scale to the upper Seine and Aube watersheds. - Mathias Maillot, Bureau de Recherches Géologiques et Minières  
16:50 - 18:00 › Modélisation des liens statistiques entre les pressions des produits phytopharmaceutiques à différentes échelles et les impacts observés dans la chimie des eaux souterraines, par apprentissage automatique - Lynh HOANG-Vy-Thuy, Bureau de Recherches Géologiques et Minières  
16:50 - 18:00 › Learning with heavy-tailed inputs: Out-of-domain generalization on extremes. - Baptiste Leroux, Mathématiques Appliquées Paris 5  

Friday, September 5, 2025

Time Event (+)
08:30 - 09:00 Welcome coffee  
09:00 - 09:40 Plenary Talk (+)  
09:00 - 09:40 › Distilling Machine Learning's Added Value: Pareto Fronts in Atmospheric Applications - Tom Beucler, University of Lausanne  
09:40 - 10:40 Deep learning (+)  
09:40 - 10:00 › Deep kernel learning for geostatistics - Thomas Romary, Mines Paris - PSL (École nationale supérieure des mines de Paris)  
10:00 - 10:20 › Score-based Generative Models for Heavy-tail and Rainfall Generation - Fassina Tiziano, Ecole des Mines - PSL - Fontainebleau  
10:20 - 10:40 › Simulations of spatial random field with deep generative adversarial networks - Charlie GARAYT, Mines Paris - PSL (École nationale supérieure des mines de Paris)  
10:40 - 11:00 Coffee break  
11:00 - 11:40 Plenary Talk (+)  
11:00 - 11:40 › Non-parametric intensity estimation of spatial point processes by random forests - Christophe Biscio, Department of Mathematical Sciences [Aalborg]  
11:40 - 12:20 Spatio-temporal (+)  
11:40 - 12:00 › Spatio-temporal generation of precipitations using a spatially correlated Bernoulli and hidden Markov model - Caroline Cognot, EDF Labs, Mathématiques et Informatique Appliquées  
12:00 - 12:20 › The modelling of counts of continuously moving individuals: a space-time geostatistical mindset for movement ecology - Ricardo Carrizo Vergara, Swiss Ornithological Institute  
12:20 - 14:00 Lunch  
14:00 - 14:40 Plenary Talk (+)  
14:00 - 14:40 › Rare event simulations, emulators, and machine learning for predicting extreme heat waves and extremes of renewable electricity production - Freddy Bouchet, Laboratoire de Météorologie Dynamique (UMR 8539)  
14:40 - 15:40 Climate and extremes (+)  
14:40 - 15:00 › Local area coefficients for spatial excursion sets, and how to use them for estimating nonstationary covariance models - Thomas Opitz, BioSP-INRAE  
15:00 - 15:20 › Graphical models for extreme events : understanding joint extremes of precipitation and river flow through graphical models - Rita Maatouk, École Nationale Supérieure des Mines de Paris [Fontainebleau]  
15:20 - 15:40 › Projecting frequencies of extreme rainfall compound events under climate change using bivariate extreme value modeling and bivariate bias corrections - Grégoire Jacquemin, Mines Paris - PSL (École nationale supérieure des mines de Paris)  
CNRS CCSD Sciencesconf