Google’s DeepMind team this week unveiled an AI model for weather forecasting called GenCast.
In an article published in NatureDeepMind researchers said they found that GenCast outperforms the European Center for Medium-Range Weather Forecasts’ ENS – apparently the world’s best operational forecasting system.
And in a blog postThe DeepMind team offered a more accessible explanation of the technology: while its previous weather model was “deterministic and provided a single, best estimate of future weather,” GenCast “includes a set of 50 or more forecasts, each representing a possible weather. trajectory,” creating a “complex probability distribution of future weather scenarios.”
As for how it compares to ENS, the team said they trained GenCast on weather data through 2018 and then compared their predictions for 2019, finding that GenCast was more accurate 97.2% of the time .
Google says GenCast is part of its suite of AI-driven weather models, which it is starting to integrate into Google Search and Maps. It also plans to release historical and real-time forecasts from GenCast, which anyone can use in their own research and models.