Research Area:  Machine Learning

Abstract:

Federated Learning (FL) is a distributed machine learning (ML) paradigm that enables multiple parties to jointly re-train a shared model without sharing their data with any other parties, offering advantages in both scale and privacy. We propose a framework to augment this collaborative model-building with per-user domain adaptation. We show that this technique improves model accuracy for all users, using both real and synthetic data, and that this improvement is much more pronounced when differential privacy bounds are imposed on the FL model.

Keywords:  
Machine Learning
Cryptography
Security
Federated Learning
Model-building

Author(s) Name:  Daniel Peterson, Pallika Kanani, Virendra J. Marathe

Publisher name:   arXiv:1912.06733

DOI:  10.48550/arXiv.1912.06733

Volume Information:  

Paper Link:   https://arxiv.org/abs/1912.06733