Federated

Federated refers to a system or arrangement in which multiple distinct entities cooperate under a common framework while retaining a degree of autonomy. In political science, a federation is a union of partially self-governing states or regions that delegate certain powers to a central authority while preserving jurisdiction over others. Constitutional arrangements define the division of powers, representation, and dispute resolution, balancing national cohesion with regional autonomy. Examples include the United States, Canada, and Germany, though federations vary in structure and level of central authority.

In information technology and digital governance, federated arrangements enable interoperability across independent domains. Federated identity management allows users to authenticate across partner organizations using a single set of credentials, typically through standards such as SAML, OAuth, or OpenID Connect, without centralizing user databases. Federated systems aim to reduce data duplication, improve user experience, and enable cross-domain collaboration while respecting domain boundaries.

Federated learning is a notable use in machine learning, where a model is trained across multiple devices or servers that hold local data. Instead of sending raw data to a central server, local models are trained and only model updates are shared and aggregated to form a global model. This approach enhances privacy and data locality but presents challenges such as communication efficiency, non-identically distributed data, and coordination overhead.

Federated search and federated services describe methods to query or operate across multiple autonomous data sources or services, presenting unified results or interfaces while preserving source autonomy.

Overall, federated approaches balance centralized coordination with distributed autonomy, trading simplicity for resilience, privacy, and domain-specific control.