Microsoft has a new machine learning tool to improve weather forecasts
Microsoft open-sourced Probabilistic Bias Correction, a CPU-friendly method that improves forecasts two to six weeks ahead.
Microsoft researchers published a Nature paper on Probabilistic Bias Correction, a machine-learning method that corrects systematic bias in existing weather models rather than training a new standalone model. In the study, PBC improved a leading ECMWF physics-based model for 98 percent of precipitation and 92 percent of temperature targets and doubled the accuracy of ECMWF’s AI forecasting system. The CPU-only tool is open source, previously won ECMWF’s AI Weather Quest, and is being tested with Kenya’s meteorological service ahead of an expected extreme rainy season.
- Probabilistic Bias Correction corrects bias in existing weather models instead of replacing them.
- It improved 98 percent of precipitation and 92 percent of temperature targets.
- The study says it doubled accuracy of ECMWF’s AI forecasting system.
- The open-source method runs on CPUs and is being tested in Kenya.
From the original article
3 min read at theverge.com
Old-school physics-based weather forecasting and even new AI models typically become less reliable beyond two weeks, a problem Microsoft claims to have solved using machine learning. Researchers with the company say they've found a new, cost-effective way to make more accurate weather forecasts two to six weeks in advance - allowing farmers and emergency responders to plan ahead with greater certainty.
By Justine Calma · theverge.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.