At a conference, an interpreter hears the noise of the hall but keeps translating. He does not repeat every word; instead, he finds an intermediate language where complex phrases become simple, yet the meaning is not lost. This is how the conditional Gaussian Koopman network works: it teaches a neural network to translate a chaotic system into a form where forecasting and data assimilation become fast and accurate.

Previously, scientists could either predict chaos with deep neural networks or assimilate observations analytically, but not both at once. Deep models were black boxes: to weave observations into them, one had to run thousands of model copies and average the results, accumulating sampling errors. The authors propose a different path: they build a neural network that itself finds an intermediate language where hidden variables become linear if the observed ones are known. Then closed-form equations for the mean and covariance of the hidden state can be written, turning data assimilation into an ordinary calculation rather than an expensive ensemble.

The interpreter who learned the dialect of chaos not only hears the noise of the hall but also instantly weaves it into the translation. He does not translate everything word for word; he finds an intermediate language where complex phrases become simple. In this language, hidden variables obey linear equations if the observed ones are known, and then data assimilation becomes analytical. This approach could accelerate climate models, geophysical forecasts, and control tasks that require fast estimates of hidden states from incomplete data.