Federated Learning is a method in the field of machine learning where many devices or computers jointly train a model without having to share their data.
Instead of collecting all data in one central location, each device keeps its own data.
The devices learn locally and only send the learned information, such as model updates, to a central server.
There, these updates are merged to improve a common model.
This way, users' privacy is protected while still benefiting from the data of all.