MIT researchers train climate emulators with synthetic data
Understand this story
MIT researchers propose a new method for training climate emulators: by running a full Earth system model just once, they can generate optimized synthetic climate scenario data for training AI climate emulators, greatly shortening the production time for such models. The research team says that the emulator, trained with only a small amount of carefully selected synthetic data, can accurately predict scenarios it has not seen during training, whereas traditional emulators cannot adapt to new situations. The related paper has been published in PNAS.
AI overview based on source reports · Updated 2 days ago
Coverage timeline
Follow the coverage to explore different perspectives.
- MIT News AINew research could lead to faster, trustworthy AI models for climate prediction
MIT researchers developed a training technique that generates optimized synthetic climate scenarios, letting a climate emulator be trained with far less data and only one full earth system model run instead of six to eight. The resulting emulator accurately predicted scenarios it had not seen during training, while traditional emulators failed to adapt, and it captured effects such as local pollutant controls amid rising other emissions. The work was published in the Proceedings of the National Academy of Sciences.
Activity for this story
- Current comparable activity
- 3
- Peak comparable activity
- 8Nov 12 22:00
- Change in the past 24 hours
- -49%
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.