Dealing with imperfect knowledge in natural hazard assessments: beyond classical probabilities and challenges
Talk, EGU 2026, Vienna (Austria)
Invited talk at EGU 2026 in the session. Download presentation here
Talks and presentations
Global Sensitivity Analysis of Spatial Interpolation for Sparse, Clustered, and Censored Data: A Case Study of Groundwater Sulfate in the Paris Basin
Talk, EGU 2026, Vienna (Austria)
Presentation of Pertchtold’s postdoc results at EGU 2026 in the session co-organised with anr HOUSES. Download presentation here
Towards harmonized operation of uncertainties in spatialized environmental systems
Talk, Challenge IA-Biodiv, Paris (France)
Presentation of HOUSES results at ANR headquarters within the final workshop of Challenge IA-Biodiv. Download presentation here
Benchmarking probabilistic spatial machine learning models with complex sample distributions
Talk, Les Journees de Geostatistique 2025, Fontainebleau
Benchmark of multiple spatial Machine Learning models for spatial interpolation with complex sample distributions (cluster, sparse) GEOSTAT2025. Download presentation here
Efficient bayesian spatially varying coefficients for censored and clustered data using the Vecchia approximation
Talk, Spatial statistics conference, Noordwijk (The Netherlands)
Developments by Y. Idir (postdoc WP3 - ARMINES) of efficient models based on bayesian spatially varying coefficients with application to soil organic carbon stock and hydrocarbon soil pollution at Toulouse (France) Download presentation here
IPMU 2024 conference
Talk, Lisboa (Portugal), Lisboa (Portugal)
Talk by Priscillia Labourg on “Geospatial uncertainties: a focus on intervals and spatial models based on inverse distance weighting” presented at the IPMU conference “20th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems”
Spatial prediction with spatially clustered data by combining residual kriging and instance-based transfer learning
Presentation, journees de la Geostatistique, Fontainebleau (France)
Presentation on the benefits of transfer learning to overcome the limits of clustered data presented at the journees de la Geostatistique. See presentation here. Download abstract here.
Explicability in Machine learning for Geoscience processes
Talk, Virtual meeting, GDR MADICS
Introduction to problems related to explicability of Machine learning models for Geoscience processes presented at the GDR MADICS. Download presentation here
Introduction Talk at ANR
Talk, Virtual meeting, ANR
Introduction to ANR-HOUSES presented at the ANR H-16 “Interface between Mathematics, Environmental and Earth sciences” axis virtual meeting. Download presentation here