BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:5th ECMWF-ESA Machine Learning Workshop
DTSTART:20260413T080000Z
DTEND:20260417T160000Z
DTSTAMP:20260812T092000Z
UID:indico-event-488@events.ecmwf.int
CONTACT:events@ecmwf.int
DESCRIPTION:\nFull size group photo\nWorkshop motivation and goals\nThe us
 e of Machine Learning (ML) technologies is becoming prevalent in an ever-g
 rowing number of applications in Earth System Observation and Prediction (
 ESOP). Additionally\, the scale\, complexity and sophistication of the ML 
 technologies applied in ESOP has also increased considerably over the last
  few years\, reflecting the growing uptake of ML ideas in the ESOP communi
 ties and benefiting from increased interest of ML domain scientists and of
  large commercial actors. This trend has reached the stage where end-to-en
 d data-driven workflows are being considered for future operational weathe
 r and climate prediction systems. On the other hand\, questions remain abo
 ut the ability of purely data driven methods to advance the science underl
 ying the slow but steady progress of ESOP capabilities based on physics-ba
 sed algorithms.  \nIn parallel with the scientific developments brought a
 bout by ML\, at the same time there is a shift in computational platforms 
 for High Performance Computing (HPC) from CPU-based to GPU/TPU-based syste
 ms. ML applications benefit from this trend as they are typically develope
 d on GPU/TPU machines\, while the process of porting codes currently used 
 for weather and climate prediction to the new computing architectures is l
 aborious. Looking further ahead\, even more disruptive technologies are ap
 pearing on the horizon\, such as quantum\, edge\, or neuromorphic computin
 g.  \nThe fifth edition of the ECMWF–ESA Workshop on Machine Learning f
 or Earth Observation and Prediction provided an up-to-date snapshot of the
  state of the art in this rapidly evolving field and a forum of discussion
  and interaction among worldwide scientists and practitioners about the cu
 rrent opportunities and challenges presented by the use of ML technologies
  for ESOP.   \nThematic areas\nPresentations focussed on all areas conne
 cted to the application of ML technologies to ESOP applications\, with a p
 articular focus on the thematic areas below.     \n\nHybrid ML-Physics 
 based systems for Data Assimilation and Weather and Climate prediction \n
 End-to-end ML systems for Data Assimilation and Weather and Climate Predic
 tion \nMachine Learning applications for Earth system observations \nHig
 h-performance and new computing technologies for ML applications in ESOP 
 \nMachine Learning for Digital Twins of the Earth system \n\n \n\nhttps:
 //events.ecmwf.int/event/488/
IMAGE;VALUE=URI:https://events.ecmwf.int/event/488/logo-2229614140.png
LOCATION:Bologna
URL:https://events.ecmwf.int/event/488/
END:VEVENT
END:VCALENDAR