If we issue too many false alarms, it reduces the trust in the forecast, but missing real events can be also dangerous.

So

Su explains what now-casting is and how her model works.

Nowcasting is a weather forecasting method that predicts conditions for the immediate future, typically in the next zero to six hours.

Nowcasting relies on real-time data from radar or satellite to track the current movement of storms.

Unlike traditional methods that would struggle with storm growth and dissipation, and also some AI models tend to blur predictions of extreme events, our deep diffusion model for satellite data uses a two-step process.

that explicitly separates the predictable large-scale motion from chaotic stochastic evolution of storm.

The first step is to use a neural network to generate deterministic baseline prediction of future satellite images.