Modern mobile devices have access to a wealth of data suitable for learning models, which is often privacy sensitive. This may preclude logging to the data center and training using conventional approaches. For this reason, here is proposed an alternative that leaves the training data distributed on the mobile devices, and learns a shared model...