Federated Learning (FL) enables mobile phones to collaboratively learnashared prediction model while keeping all the training data on device, decoupling the ability to do machine learning from the need to store the data in the cloud. This goes beyond the use of local models that make predictions on mobile devices by bringing model training to the device as well. FL was co-invented by my former student Jakub Konecny, myself and Google.
We have launched. a FL system in 2017, it is now in use in more than 1 billion Android devices:
In this project we will investigate further improvements and applications of FL.