Federated learning trains a shared model across many devices without collecting their data on a server. That is a real and useful property. It is also routinely stated as though it were three properties, and the other two — that the server learns nothing about individuals, and that the resulting model cannot leak training data — are not true unless you add machinery that costs something.
The protocol, step by step
The canonical algorithm is federated averaging. One round looks like this:
The server selects a cohort of eligible clients — typically devices that are charging, idle and on an unmetered connection, because training is expensive and users must not notice.
The server sends each selected client the current global model parameters.
