A Systematic Literature Review on Federated Machine Learning: From A Software Engineering Perspective

By Sin Kit Lo et al

Published on Aug. 10, 2020

Read the original document by opening this link in a new tab.

Table of Contents

1. Introduction
2. Methodology
3. Results

Summary

This paper presents a systematic literature review on federated machine learning from a software engineering perspective. It covers the lifecycle of federated learning system development, including background understanding, requirement analysis, architecture design, implementation, and evaluation. The study synthesizes data from 231 primary studies and identifies future trends in federated learning research. The research questions focus on various aspects such as training settings, data distribution, orchestration, client types, and data partitioning in federated learning systems. The paper provides insights into the challenges and approaches in developing federated learning systems.