Data Science In Context
Stephanie Roark
12/05/2018
What is Federated Learning?
Federated Learning is machine learning where training data is distributed over a large number of clients each with unreliable and relatively slow network connections. The updates from these clients are used to train a high-quality centralized model which is then re-distributed to the clients.
Typical clients are mobile devices
Most Machine Learning happens in the cloud after collecting data and transferring to the cloud for training.
How to centralize modeling without high data transfer costs and while maintaining privacy?
Jesus Rodriguez, 02-26-18, What's New in Deep Learning Research, https://towardsdatascience.com/whats-new-in-deep-learning-research-understanding-federated-learning-b14e7c3c6f89
Gabriel de Vinzelles, 03-06-18, Federated Learning, a step closer towards confidential AI, https://hackernoon.com/federated-learning-a-step-closer-towards-confidential-ai-7ac4afa9b437
Alex Ingerman, https://www.youtube.com/watch?v=UgiPrYhBYYo&feature=youtu.be
Robin C. Geyer, Tassilo Klein, Moin Nabi, 12-20-17, Differentially Private Federated Learning: A Client Level Perspective, https://arxiv.org/abs/1712.07557
Keith Bonawitz and Vladimir Ivanov and Ben Kreuter and Antonio Marcedone and H. Brendan McMahan and Sarvar Patel and Daniel Ramage and Aaron Segal and Karn Seth, 2016, Practical Secure Aggregation for Federated Learning on User-Held Data, https://ai.google/research/pubs/pub45808
Brendan McMahan and Daniel Ramage, Research Scientists, 04-06-17, Federated Learning: Collaborative Machine Learning without Centralized Training Data, https://ai.googleblog.com/2017/04/federated-learning-collaborative.html