FEDERATED LEARNING . noun . \ ˈfe-də-ˌrā-təd \ ˈlər-niŋ
(Ok, this one is complicated. We're going to simplify it best we can!)
: Federated Learning (FL) is a machine learning approach that downloads the algorithm, or model, at the device itself (where the data is collected and stored).
The algorithm is then applied to the local data and the results (not the data!) are sent back to a central server where they are consolidated and analyzed according to needs.
🙌 The major benefit of Federated Learning is DATA SECURITY. 🙌
Instead of transferring data to a central server to be analyzed, FL sends the analyzation tool (the algorithm or model) to the data! Only the analyzed results will be moved around, protecting the security and ANONYMITY of the individual data!
If you'd like to dive deeper into Federated Learning, Google has a couple great articles.
For more technical information, this article is very informative.
Or, if you're looking for more humor and color, then check out this article here.
Any further questions? Don't hesitate to reach out! Federated Learning is APURBA Ltd's specialty and we'd love to tell you more.