Back to Glossary
AdvancedAI Glossary

Federated Learning

Quick Answer

Federated learning is an approach to training machine learning models across multiple decentralised devices or servers, without moving the underlying data to a central location. Only model updates are exchanged, preserving privacy and data sovereignty.

In Depth

What Federated Learning really means

Federated learning is attractive wherever data is sensitive or legally restricted: across hospitals sharing imaging models, banks collaborating on fraud detection, or mobile devices improving on-device keyboards.

Challenges include non-identically distributed data across participants, communication overhead, and the need for robust protocols to prevent information leakage through model updates themselves.

Why It Matters

Business relevance for UK organisations

UK healthcare, financial services and research organisations increasingly explore federated learning to collaborate on model quality without breaching data-protection obligations.

Real-world example

How this shows up in practice

A consortium of UK NHS trusts used federated learning to train a radiology triage model across hospitals, without any patient images leaving individual hospital networks.