Federated learning is increasingly practical for machine learning developers because of the challenges we face with model and data privacy. In this fully connected episode, Chris and Daniel dive into the topic and dissect the ideas behind federated learning, practicalities of implementing decentralized training, and current uses of the technique.