AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
How Do Neural Operators Compare to Traditional Models?
With these models, you've referred to the relative speed up. In the case of the weather, we are doing as well as the current weather models for as much as two weeks - which is considered where it's very predictive. So after that and a bit longer, you kind of get to the sub-seasonal and seasonal scales where it's no longer predictable. So it becomes chaotic. You can kind of say statistical measures on average, what it would look like, but not the actual trajectory,. the actual weather, what it is because it's just not predictive.