Microsoft has a new machine learning tool to improve weather forecasts

Weather models fill gaps in sparse data with approximations, so errors compound in forecasts beyond two weeks. Microsoft's new Probabilistic Bias Correction uses machine learning to find and fix a model's systematic bias: it improved a leading weather model for 98 percent of precipitation targets and doubled its AI system's accuracy. The tool is open source and runs on CPUs, which helps smaller agencies.

Summary written with AI by HowTechMoves, based on reporting by The Verge.