When Helene made landfall in Florida earlier this year, 234 people lost their lives in the worst hurricane to hit the continental US since Katarina in 2005. It's natural disasters like that, and their increasing intensity due to changing climate, which have driven scientists to develop more accurate weather forecasting systems. On Wednesday, Google's DeepMind division announced what may be considered the most significant advance in the field in nearly eight decades of work.
In a publication in the Google Keywords BlogDeepMind's Ilan Price and Matthew Wilson detailed GenCast, the company's latest ai agent. According to DeepMind, GenCast is not only better at providing daily and extreme weather forecasts than its previous ai weather program, but it also outperforms the best forecasting system currently in use, one maintained by the European Center for Medium Weather Forecasts. Term (CEMPM). In tests comparing the 15-day forecasts the two systems generated for the weather in 2019, GenCast was, on average, more accurate than ECMWF's ENS system 97.2 percent of the time. With turnaround times exceeding 36 hours, DeepMind's was even better, 99.8 percent more accurate.
“I'm a little reluctant to say it, but it's like we've made decades of improvements in one year,” said Rémi Lam, lead scientist at DeepMind's former ai weather program. <a target="_blank" data-i13n="cpos:2;pos:1" href="https://www.nytimes.com/2024/12/04/science/google-ai-weather-forecast.html” rel=”nofollow noopener” target=”_blank” data-ylk=”slk:told The New York Times;cpos:2;pos:1;elm:context_link;itc:0;sec:content-canvas” class=”link “>said The New York Times. “We're seeing really rapid progress.”
GenCast is a broadcast model, which is the same technology that powers Google's generative ai tools. DeepMind trained the software with almost 40 years of high-quality weather data curated by the European Center for Medium-Range Weather Forecasts. The predictions generated by the new model are probabilistic, meaning they take into account a range of possibilities that are then expressed as percentages. Probabilistic models are considered more nuanced and useful than their deterministic counterparts, which only offer a better estimate of what the weather might be like on a given day. The former are also more difficult to create and calculate.
In fact, what is perhaps most surprising about GenCast is that it requires significantly less computing power than traditional physics-based ensemble forecasts like ENS. According to Google, a single of its TPU v5 Tensor Processing Units can produce a 15-day GenCast forecast in eight minutes. By contrast, a supercomputer with tens of thousands of processor hours can take time to produce a physics-based forecast.
Of course, GenCast isn't perfect. One area where the software could provide better predictions is hurricane intensity, although the DeepMind team said The times He was confident that he could find solutions to the agent's current deficiencies. Meanwhile, Google is making GenCast an open model, with example code for the tool. available on GitHub. GenCast predictions are also coming soon to Google Earth.
