Improved Information Theoretic Generalization Bounds for Distributed and Federated Learning

A Google TechTalk, presented by Leighton Pate Barnes, Princeton University, at the 2021 Google Federated Learning and Analytics Workshop, Nov. 8-10, 2021. Full title: Improved Information Theoretic Generalization Bounds for Distributed and Federated Learning with Model Averaging For more information about the workshop: https://events.withgoogle.com/2021-workshop-on-federated-learning-and-analytics/#content