AI-enhanced rare-event sampling helps predict extreme weather
Physics World
Researchers have developed an AI-boosted rare-event sampling framework that significantly improves the accuracy and efficiency of predicting extreme weather events like heatwaves.Summary
Researchers from the US and France have developed an AI-enhanced rare-event sampling (AI+RES) framework to improve the prediction of extreme climate events. Traditional climate models require excessive computational power to simulate rare occurrences, while pure AI models often lack physical reliability and insight. The new method uses AI to identify conditions likely to trigger extreme events, allowing physical climate models to focus computational resources specifically on those instances. Testing against the PlaSim model showed the framework achieved similar accuracy with up to 1000 times fewer computational resources, offering a promising, scalable tool for future climate science applications.
(Source：Physics World)