 is a technique for training machine learning models without moving large amounts of data to a central server

Instead, copies of the model are sent to the devices where the data resides, and the model is trained locally on each device

The updated models are then sent back to a central server, where they are aggregated to improve the global model without revealing any private data. 

it used for applications such as improving word recommendation on 

The cost for implementing federated learning is higher than collecting the information and processing it centrally, especially during the early phases of R&D when the training method and process are still being iterated on

So the concept of importing something that is considered to some extent to be the default (