Lost (and found) in latent land
AI and machine learning boost weather and climate models’ accuracy. Analysing latent spaces reveals new insights on water, carbon, and atmospheric processes, improving model understanding.
-
Date: 14 November 2025 from 16:30 to 17:30
-
Event location: AULA 4 - FISICA Piano Primo Interrato viale Berti Pichat 6-6/2 BOLOGNA
-
Access Details: Free admission
Hybrid event
Machine learning and artificial intelligence have been revolutionising weather and climate modelling and data analysis over the past few years. However, it remains unclear how much understanding has been gained from those models, even though they are reaching unprecedented accuracy. Through the study and analysis of carefully chosen latent spaces, I will demonstrate how we can get new understanding on the terrestrial water and carbon cycle as well as on atmospheric processes, specifically on convection. Those latent spaces can also be used to better characterise complex stochastic processes, such as turbulence, and combined with data assimilation in order to achieve improved performance in those models.
SPEAKER
Pierre Gentine
Department of Earth and Environmental Engineering & Department of Earth and Environmental Sciences, Columbia University, USA
PhD from MIT in 2010. Faculty of the Dept of Applied Mathematics and Applied Physics at Columbia from 2010. Recipient of NASA, DOE, NSF Early Career Award and AGU Macelwane medalist.