COMPEL Glossary / federated-learning
Federated Learning
Federated learning is a machine learning approach where a model is trained across multiple devices, servers, or organizations holding local data, without exchanging the raw data itself.
What this means in practice
Instead, each participant trains a local model on their data and shares only the model updates (gradients or parameters) with a central coordinator that aggregates them into a global model. This enables collaborative AI development while keeping sensitive data in place. Federated learning is particularly valuable in healthcare (hospitals can collaborate on diagnostic AI without sharing patient records), financial services (banks can build fraud models without sharing transaction data), and enterprise settings where data cannot leave specific jurisdictions due to regulations. Federated learning is an emerging technology assessed in the COMPEL Technology pillar as organizations explore privacy-preserving AI approaches.
Why it matters
Federated learning enables collaborative AI development across organizations without sharing raw data, addressing critical barriers in industries where data cannot be centralized due to privacy regulations, competitive concerns, or sovereignty requirements. Healthcare organizations can collaborate on diagnostic AI without sharing patient records, and financial institutions can build fraud models without exchanging transaction data, unlocking value that would otherwise remain inaccessible.
How COMPEL uses it
Federated learning is assessed in the COMPEL Technology pillar as organizations explore privacy-preserving AI approaches. During Calibrate, the need for federated approaches is evaluated based on data sensitivity and collaboration requirements. During Model, federated learning architecture decisions are made as part of the Technology pillar design. Module 4.3 covers cross-organizational governance requirements that federated learning deployments must satisfy.
Related Terms
Other glossary terms mentioned in this entry's definition and context.