[IDA ML Seminar] Machine Learning Seminar, 20/3 at 15:15: Markus Heinonen, "Can neural ODEs forecast global weather?"

[Fredrik Lindsten]

Hi all,

A gentle reminder about the ML seminar today at 15:15 in Ada Lovelace. If you are unable to attend in person, it is also possible to join over Zoom. See details below.

Best,
Fredrik


Fredrik Lindsten (LiU) is inviting you to a scheduled Zoom meeting.

Topic: ML Seminar: Markus Heinonen
Time: Mar 20, 2024 03:15 PM Stockholm

Join Zoom Meeting
https://liu-se.zoom.us/j/68178052431?pwd=cS9uVFRtYTlTMERMVjJqY1pNNlgwUT09

Meeting ID: 681 7805 2431
Passcode: 058839

---

One tap mobile
+46850539728,,68178052431#,,,,*058839# Sweden
+46844682488,,68178052431#,,,,*058839# Sweden

---

Dial by your location
* +46 850 539 728 Sweden
* +46 8 4468 2488 Sweden

Meeting ID: 681 7805 2431
Passcode: 058839

Find your local number: https://liu-se.zoom.us/u/ckFlB8S4R

---

Join by SIP
68178052431 at 109.105.112.236<mailto:68178052431 at 109.105.112.236>
* 68178052431 at 109.105.112.235<mailto:68178052431 at 109.105.112.235>

---

Join by H.323
* 109.105.112.236
* 109.105.112.235

Meeting ID: 681 7805 2431
Passcode: 058839






From: Fredrik Lindsten
Sent: den 13 mars 2024 19:02
To: ml-seminars at lists.liu.se; idaint at ida.liu.se; elliit at lists.liu.se; ai-academy at groups.liu.se
Subject: Machine Learning Seminar, 20/3 at 15:15: Markus Heinonen, "Can neural ODEs forecast global weather?"

Welcome to an IDA Machine Learning Seminar on Wednesday, March 20 at 15:15 in Ada Lovelace

Can neural ODEs forecast global weather?
Markus Heinonen<https://users.aalto.fi/~heinom10/>, Department of Computer Science, Aalto University
Abstract: Neural ODEs have surfaced in the last decade as a new perspective on modelling dynamics by learning the time-derivative that drives the system evolution forward as a neural network. While successful on systems of limited complexity, large-scale demonstrations have been lacking. Recently large autoregressive transformer models have been developed with breakthrough global weather forecasting performances, but at times with little consideration of the underlying dynamics. We consider global weather as a continuous-time PDE with mass-preserving dynamics, and show how simple convolution networks can achieve state-of-the-art weather prediction performance with just a few million parameters. This talk is based on the paper ClimODE: Climate Forecasting With Physics-informed Neural ODEs<https://openreview.net/forum?id=xuY33XhEGR> accepted for ICLR24 oral presentation.
Bio: Markus Heinonen is an Academy Research Fellow at Aalto University, Finland with a PhD from University of Helsinki in 2013. His research interests are centered on probabilistic machine learning with emphasis on understanding uncertainty of deep learning with Bayesian perspectives to neural networks. In addition he has worked on learning ODEs

Location: Ada Lovelace, https://www.ida.liu.se/department/location/search.en.shtml?keyword=ada

------------------

The list of future seminars in the series is available at: http://www.ida.liu.se/research/machinelearning/seminars/
You can subscribe to the seminar series' calendar using this ics link: https://outlook.office365.com/owa/calendar/4d811ae47ce446f58d11a7c2f50a7ed8@ad.liu.se/0f5253d7bc7841248c71eb4c28eb2d668927992292494627279/calendar.ics





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.liu.se/pipermail/ml-seminars/attachments/20240320/7fdec710/attachment.htm>