BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//juliacon-2022//speaker//A9SQSW
BEGIN:VEVENT
UID:pretalx-juliacon-2022-DZFPGX@pretalx.com
DTSTART:20220728T171000Z
DTEND:20220728T172000Z
DESCRIPTION:We present SpeedyWeather.jl\, a global atmospheric model curren
 tly developed as a prototype for a 16-bit climate model incorporating mach
 ine learning for accuracy and computational efficiency on different hardwa
 re. SpeedyWeather.jl is designed for type flexibility with low precision\,
  and automatic differentiation to replace parts of the model with neural n
 etworks for a more accurate representation of climate processes and comput
 ational efficiency.
DTSTAMP:20260308T035456Z
LOCATION:Blue
SUMMARY:SpeedyWeather.jl: A 16-bit weather model with machine learning - Mi
 lan Klöwer
URL:https://pretalx.com/juliacon-2022/talk/DZFPGX/
END:VEVENT
END:VCALENDAR