AI Glossary

Federated learning

What does “Federated learning” mean?

Federated learning: Training across separate devices or organizations by sharing updates under a coordination protocol.

Federated learning visual guide: Training across separate devices or organizations by sharing updates under a coordination protocol.
Visual guide Open full size (new tab)

What is an example of “Federated learning”?

Phones train a shared model while keeping their raw keyboard examples on the device.

Keep learning

See also

DatasetTermTraining dataTermData labelingTermWeak supervisionTermActive learningTerm

Same topic

Domain adaptationTermTransfer datasetTermFine-tuningTermExploration and exploitationTermGradient accumulationTerm