Federated learning is a machine learning approach where models are trained across multiple distributed data sources without centralizing the data in one location. Each participating node trains on its local data and shares only model updates — not the raw data itself — with a central aggregator.
It is particularly valuable when dealing with sensitive data that cannot be shared across organizational or regulatory boundaries, such as patient records across hospitals or financial data across banks. Federated learning makes it possible to train on data that could never be pooled in one place, while still producing a model that benefits from all of it.
