BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Gustau Camps-Valls (Universitat de València) DTSTART:20210528T130000Z DTEND:20210528T140000Z DTSTAMP:20260423T003240Z UID:MPML/45 DESCRIPTION:Title: Ph ysics Aware Machine Learning for the Earth Sciences\nby Gustau Camps-V alls (Universitat de València) as part of Mathematics\, Physics and Machi ne Learning (IST\, Lisbon)\n\n\nAbstract\nMost problems in Earth sciences aim to do inferences about the system\, where accurate predictions are jus t a tiny part of the whole problem. Inferences mean understanding variable s relations\, deriving models that are physically interpretable\, that are simple parsimonious\, and mathematically tractable. Machine learning mode ls alone are excellent approximators\, but very often do not respect the m ost elementary laws of physics\, like mass or energy conservation\, so con sistency and confidence are compromised. I will review the main challenges ahead in the field\, and introduce several ways to live in the Physics an d machine learning interplay that allows us (1) to encode differential equ ations from data\, (2) constrain data-driven models with physics-priors an d dependence constraints\, (3) improve parameterizations\, (4) emulate phy sical models\, and (5) blend data-driven and process-based models. This is a collective long-term AI agenda towards developing and applying algorith ms capable of discovering knowledge in the Earth system.\n LOCATION:https://researchseminars.org/talk/MPML/45/ END:VEVENT END:VCALENDAR