Federated Evaluation of On-device Personalization

Abstract

Federated learning is a distributed, on-device computation framework that enables training global models without exporting sensitive user data to servers. In this work, we describe methods to extend the federation framework to evaluate strategies for personalization of global models. We present tools to analyze the effects of personalization and evaluate conditions under which personalization yields desirable models. We report on our experiments personalizing a language model for a virtual keyboard for smartphones with a population of tens of millions of users. We show that a significant fraction of users benefit from personalization.

4 pages, 4 figures

Cited by in corpus (4)

• Federated Learning for Healthcare Domain - Pipeline, Applications and Challenges
• Implicit Model Specialization through DAG-based Decentralized Federated Learning
• FedPref: Federated Learning Across Heterogeneous Multi-objective Preferences
• MultiConfederated Learning: Inclusive Non-IID Data handling with Decentralized Federated Learning