Programme

Download Programme
The times below are displayed for Europe/London. We have detected that you are visiting us from .

Please select a timezone to show adjusted event times accordingly.



Wednesday

The time displays according to the selected time zone.

Wednesday, 27 September 2023
Welcome and introductions
09:00 to 09:30 Welcome and introduction to ECMWF-MS ML project
Matthew Chantry (ECMWF)
09:30 to 10:30 Tour-de-table
10:30 to 11:00 Group photo / Coffee break
11:00 to 12:00 Tour-de-table (continued)
12:00 to 12:30 Discussion on datasets
12:30 to 14:00 Lunch break
Theme: Data-driven forecasting
14:00 to 14:30 Evaluation of open-source data-driven models
Zied Ben Bouallegue (ECMWF)
14:30 to 15:00 The Moonshot Project: towards multi-source data-driven forecasting
Tom Dunstan (Met Office)
15:00 to 15:30 AIFS
Presentation slides
Simon Lang (ECMWF)
15:30 to 16:00 Coffee break
16:00 to 16:30 GraphCast for a limited area numerical weather prediction model
Tomas Landelius (SMHI)
16:30 to 17:30 Discussion of data-driven forecasts
17:30 to 19:30 Drinks and tapas reception

Weather Room


Thursday

The time displays according to the selected time zone.

Thursday, 28 September 2023
Theme: Model enhancement
09:00 to 09:30 Hybrid NWP-ML forecasting
Matthieu Plu (Météo-France)
09:30 to 10:00 Machine learning methods and its challenges in post-processing at Met Norway
John Bjørnar Bremnes (Norwegian Meteorological Institute)
10:00 to 10:30 Discussion on model enhancement
10:30 to 11:00 Coffee break
Theme: Computing and Foundation models
11:00 to 11:30 Discussion on computing
11:30 to 12:00 Foundation models
Christian Lessig (ECMWF)
12:00 to 12:30 Discussion on foundation models
12:30 to 13:45 Lunch break
Theme: MLOps
13:45 to 14:15 Machine learning activities at MeteoSwiss
Lionel Moret (MeteoSwiss)
14:15 to 14:45 Discussion on MLOps
14:45 to 15:00 Coffee break
15:00 to 16:00 Summary discussions