Microsoft open-sources weather bias correction method PBC
What happened
Reported on October 15, Microsoft Research introduced Probabilistic Bias Correction (PBC), a machine learning method that corrects systematic bias in existing weather models to improve forecasts two to six weeks out. The report says PBC raised the accuracy of the ECMWF physical model on 98% of precipitation targets and 92% of temperature targets, and doubled the accuracy of its AI forecasting system. The tool is open source, runs on CPU alone, and the research was published in Nature.
Written by AI from the coverage · updated 1 hour ago
Coverage
Follow the reports to see the story from different sides.
- The Verge · AIMicrosoft has a new machine learning tool to improve weather forecasts
微软研究院推出机器学习方法 Probabilistic Bias Correction（PBC），通过修正现有天气模型的系统性偏差，将 ECMWF 物理模型在 98% 降水目标和 92% 温度目标上的准确率提升，并使其 AI 预报系统准确率翻倍。该工具已开源，仅需 CPU 即可运行，相关成果发表于 Nature。
Heat over time
Not enough continuous observations to draw a trend yet.