BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20190719T085745Z
LOCATION:HG D 1.2
DTSTART;TZID=Europe/Stockholm:20190614T150000
DTEND;TZID=Europe/Stockholm:20190614T153000
UID:submissions.pasc-conference.org_PASC19_sess114_msa198@linklings.com
SUMMARY:Neural Networks for Weather and Climate Forecasting: Training on P
 ast Weather Forecasts and Output of Simple Climate Models
DESCRIPTION:Minisymposium\nClimate and Weather\n\nNeural Networks for Weat
 her and Climate Forecasting: Training on Past Weather Forecasts and Output
  of Simple Climate Models\n\nScher, Messori\n\nWe present progress on usin
 g artificial neural networks, specifically Convolutional Neural Networks (
 CNNs) for the problem of weather and climate forecasting. First the use of
  CNNs to predict the uncertainty of synoptic weather forecasts, trained on
  past weather forecasts, will be presented. Then we will move on to predic
 t the model state of a range of simple general circulation models several 
 days ahead, thus effectively making “weather” forecasts in the
  simplified reality of the model. Additionally, we assess the possibility 
 of using these trained networks in order to make very long forecasts, thus
  creating a climate-timeseries that emulates the general circulation model
 . Finally, the potential to applying these techniques to real-world weathe
 r forecasting will be discussed.
END:VEVENT
END:VCALENDAR