Benjamin Carr, Ph.D. 👨🏻💻🧬<p><a href="https://hachyderm.io/tags/Google" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Google</span></a>’s <a href="https://hachyderm.io/tags/DeepMind" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>DeepMind</span></a> tackles <a href="https://hachyderm.io/tags/weather" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>weather</span></a> forecasting, with great performance<br>Google's DeepMind claims its new <a href="https://hachyderm.io/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> system manages to outperform European model on <a href="https://hachyderm.io/tags/forecasts" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>forecasts</span></a> out to at least a week and often beyond. DeepMind's system, called <a href="https://hachyderm.io/tags/GenCast" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>GenCast</span></a>, merges some computational approaches used by atmospheric scientists with a diffusion model, commonly used in generative AI. The result is a system that maintains high resolution while cutting the computational cost significantly. <br><a href="https://arstechnica.com/science/2024/12/googles-deepmind-tackles-weather-forecasting-with-great-performance/" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">arstechnica.com/science/2024/1</span><span class="invisible">2/googles-deepmind-tackles-weather-forecasting-with-great-performance/</span></a></p>