Oliver Watt-Meyer

93 posts

Oliver Watt-Meyer

Oliver Watt-Meyer

@oliwm

Lead Research Scientist for Climate Modeling @allen_ai | Associate Editor @theAGU JAMES | Opinions my own

Seattle, WA Katılım Haziran 2016
195 Takip Edilen419 Takipçiler
Sabitlenmiş Tweet
Oliver Watt-Meyer
Oliver Watt-Meyer@oliwm·
🌎 We developed an ML-based atmospheric model suitable for climate prediction! Long-term stable, capable of assessing mass/moisture conservation, and predicts diagnostics for (eventually) coupling to ocean. Very excited about this project! 🎉
Ai2 Climate Modeling@ai2_climate

We're excited to present ACE (AI2 Climate Emulator), a long-term (at least 10 yr) stable ML emulator of a comprehensive 100-km resolution global atmospheric model: arxiv.org/abs/2310.02074

English
2
5
41
4.5K
Oliver Watt-Meyer retweetledi
Ai2
Ai2@allen_ai·
🌍☀️❄️ Can AI forecast year-to-year differences in the seasons? New research with @metofficeUK shows our ACE2 ML model demonstrates seasonal forecasting skill—matching traditional physics-based methods while using dramatically less compute. 🧵
Ai2 tweet media
English
4
3
17
3.1K
Oliver Watt-Meyer retweetledi
Ai2 Climate Modeling
Ai2 Climate Modeling@ai2_climate·
We are now seeking applicants for a full-time research scientist position on the climate modeling team at Ai2!  Come work with an excellent team at the frontier of using AI for climate modeling.  🌎🌦 Hybrid in Seattle, app review begins Oct 15. Details 👉job-boards.greenhouse.io/thealleninstit…
English
0
3
9
737
Oliver Watt-Meyer retweetledi
Pierre Gentine
Pierre Gentine@PierreGentine·
Excited to share our new review paper on the use of ML for climate science rdcu.be/dRMdh led by Veronika Eyring and Bill Collins and based on a great workshop in Colorado in 2022
English
0
11
83
15.3K
Oliver Watt-Meyer retweetledi
ml4esmworkshop
ml4esmworkshop@ml4esmworkshop·
We are delighted to announce the #ICML2024 ML4ESM Best Paper Award winner: [2406.14798] Probabilistic Emulation of a Global Climate Model with Spherical DYffusion (arxiv.org) Congrats to @salvaRC7 and co-authors @ai2_climate @yuqirose for their amazing work!
English
0
12
61
40K
Oliver Watt-Meyer retweetledi
ml4esmworkshop
ml4esmworkshop@ml4esmworkshop·
Hope everyone is enjoying #ICML2024! Don't forget to stop by our workshop on Machine Learning for Earth System Modeling tomorrow, Friday 26th in Stolz 1! Schedule here: icml.cc/virtual/2024/w…
English
1
5
7
1.9K
Oliver Watt-Meyer retweetledi
Ai2
Ai2@allen_ai·
How can AI help us prepare for a changing climate? That's one of the big questions our climate modeling senior director Chris Bretherton wants to answer. Chris spoke with @nvidia's @noahkravitz about how technology like localized models can help us prepare for the future — listen here: blogs.nvidia.com/ai-podcast/
English
3
2
14
4.8K
Oliver Watt-Meyer retweetledi
Ai2 Climate Modeling
Ai2 Climate Modeling@ai2_climate·
We have a new ACE preprint led by our former intern, James Duncan, where we’ve successfully emulated a second major atmospheric model, DOE’s EAMv2! 🌎💻 doi.org/10.22541/au.17…
Ai2 Climate Modeling tweet media
English
2
4
15
2.3K
Oliver Watt-Meyer retweetledi
Blanka Balogh (@blankabalogh.bsky.social)
Did you like the IA NWP models? You'll love the AI climate models! The authors present a climate model based on SFNO block, and with an impressive numerical stability of 10 years (!) at least. arxiv.org/abs/2310.02074
English
1
3
23
1.6K
Oliver Watt-Meyer retweetledi
Ai2
Ai2@allen_ai·
It was a breakthrough year for environmental AI, with many new foundation models and innovative solutions for tackling our biggest environmental challenges. Learn more about 2023's major advances and key themes in climate and sustainability research: blog.allenai.org/ai-for-earth-2…
English
0
4
15
3.7K
Oliver Watt-Meyer
Oliver Watt-Meyer@oliwm·
@janniyuval For stability, choice in architecture, input variables (insolation, surface height), and normalization strategy all led to improvements. Our dataset on gaussian grid is also well suited for the architecture we used. Haven't done enough ablations to say exactly which matters most.
English
0
0
1
56
janniyuval
janniyuval@janniyuval·
@oliwm Thanks for sharing! Could you highlight what improved long term stability? Also, studies that used MSE loss had blurring. Do you have blurring?. A thought I had - blurring might not be a problem because of the ground truth that is more blurry than ERA5?
English
2
0
2
103
Oliver Watt-Meyer
Oliver Watt-Meyer@oliwm·
🌎 We developed an ML-based atmospheric model suitable for climate prediction! Long-term stable, capable of assessing mass/moisture conservation, and predicts diagnostics for (eventually) coupling to ocean. Very excited about this project! 🎉
Ai2 Climate Modeling@ai2_climate

We're excited to present ACE (AI2 Climate Emulator), a long-term (at least 10 yr) stable ML emulator of a comprehensive 100-km resolution global atmospheric model: arxiv.org/abs/2310.02074

English
2
5
41
4.5K
Oliver Watt-Meyer
Oliver Watt-Meyer@oliwm·
@janniyuval We have some blurring but less than what I've seen for models using 0.25° ERA5. Using 1° data makes it easier. But also we only optimize over a single 6hr step which makes a big difference.
English
0
0
2
76
Stephan Hoyer
Stephan Hoyer@shoyer·
@oliwm Amazing stuff! It's awesome to see pure AI emulators working for climate. Let's definitely catch up at AGU :)
English
1
0
0
369
Oliver Watt-Meyer
Oliver Watt-Meyer@oliwm·
I'll be at @theAGU fall meeting presenting this next week—look forward to chatting with others there. Lots to catch up on.
English
1
0
0
179