IT Home May 24 NewsparallelblockchaingameThe Microsoft Start team posted a blog post today through the Bing official blog, announcing that the AI weather prediction model has been significantly improved. In addition to more accurately predicting weather conditions in the next 30 days, it is now predicting when clouds and precipitation may appear.
In the blog post, the Microsoft Start team said that it will start at the end of 2021 and start to improve global performanceparallelblockchaingameThe "precipitation forecast" system.
This model not only calls data collected by radar devices, but also combines satellite data and analyzes it through AI algorithm to more accurately predict precipitation.
However, the Microsoft Start team also admitted that there are shortcomings in the prediction model. The daily reception time of satellite weather data is 85% to 95%, depending on the source of the data and acceptable delay time.
Microsoft said:
There is evidence that each task requires a separate decoder and each prediction channel requires a separate discriminator, so the model built by Weather from Microsoft Start is four times larger than previous models that only predicted simulated radar reflectivity. Finally, the new model jointly predicts satellite and simulated radar reflectivity, allowing its predictions to fill gaps in data availability.
According to Microsoft, in artificial intelligence training, radar channel models are six times more weighted than satellite models because precipitation information provided by radar is considered more important than satellite missions. The net result is that the new precipitation and cloud model "shows significant improvements in F1 scores" compared to the basic model that uses only radar predictions.
IT Home Attach reference address