Research on Privacy and Security Issues in Federated Learning

Xinyuan Bi

2025

Abstract

In the digital age, data privacy and security have become key issues. Federated learning, as an emerging distributed machine learning technology, has been widely applied in fields such as finance and healthcare. However, federated learning still faces many challenges in privacy protection. This paper deeply studies the privacy and security issues of federated learning analyzes its technical principles, privacy protection mechanisms, and performance in practical applications. Through the analysis of application cases in fields such as finance and healthcare, it explores the advantages and disadvantages of federated learning in privacy protection. On this basis, this paper proposes improvement strategies such as optimizing encryption technology, strengthening model security, establishing a security audit mechanism, and improving laws and regulations, aiming to enhance the privacy and security level of federated learning and provide guarantees for its stable application in various fields. In the future, the research on privacy and security of federated learning will develop towards more intelligent, efficient, and integrated directions. It is necessary to further study new privacy protection technologies, strengthen dynamic security monitoring and adaptive defense capabilities, and formulate unified privacy and security standards and norms to promote the safe application and development of federated learning technology worldwide.

Paper Citation

in Harvard Style

Bi X. (2025). Research on Privacy and Security Issues in Federated Learning. In Proceedings of the 2nd International Conference on Data Science and Engineering - Volume 1: ICDSE; ISBN 978-989-758-765-8, SciTePress, pages 104-110. DOI: 10.5220/0013679400004670

in Bibtex Style

@conference{icdse25,
author={Xinyuan Bi},
title={Research on Privacy and Security Issues in Federated Learning},
booktitle={Proceedings of the 2nd International Conference on Data Science and Engineering - Volume 1: ICDSE},
year={2025},
pages={104-110},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013679400004670},
isbn={978-989-758-765-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Data Science and Engineering - Volume 1: ICDSE
TI - Research on Privacy and Security Issues in Federated Learning
SN - 978-989-758-765-8
AU - Bi X.
PY - 2025
SP - 104
EP - 110
DO - 10.5220/0013679400004670
PB - SciTePress