BEGIN:VCALENDAR
VERSION:2.0
X-WR-CALNAME;VALUE=TEXT:ClimaTea Journal Club
PRODID:-//Harvard events data//EN
BEGIN:VEVENT
UID:event_1326387_0
SUMMARY:ClimaTea Journal Club
DESCRIPTION:<p>	Speaker: Packard Chan</p><p>	We are glad to have Packard to discuss machine learning in climate models. Please see the attached paper (<a href="http://www.pnas.org/content/early/2018/09/05/1810286115">link</a>).</p><p>	During he discussion, Packard will introduce the following points:<br>1. Deep learning can represent subgrid processes in climate models.</p><p>	<span><span>2. The prognostic multiyear simulations are stable and closely reproduce mean climate and variabilities.<br>3. Generalization works in some cases but fails the others.</span></span></p>
LOCATION:HUCE Seminar Room MCZ 440
STATUS:CONFIRMED
DTSTART:20180925T190000Z
DTEND:20180925T190000Z
END:VEVENT
END:VCALENDAR