Skip to main content

Post from datos.gob.es

· One min read
ClĂ udia ColĂ s Torres
Estudiante de la Universitat de Lleida

📌 Very interesting post from datos.gob.es - Data Office on the role of federated learning

Data spaces allow data to be reused without moving it, thanks to federated computing, which generates intermediate results while preserving data privacy and control.

One of its most notable applications is federated learning, which allows AI models to be trained without centralizing data, thus reinforcing data sovereignty.

💡 In the AgrospAI data space demonstrator, we apply this approach through Compute-to-Data: data is not copied, but rather services "visit" the data at its source. This way, providers maintain control over their data, preventing the dissemination of personal or sensitive data.

Through data sovereignty by design, it's possible to create a trusted environment that allows us to add value to the data we need to keep under control. You can read more in this article.

#DataSpaces #EspaciosDeDatos #FederatedLearning #ComputaciĂłnFederada #SoberanĂ­aDelDato #ComputaciĂłnADatos