An Alternative Approach to Federated Learning for Model Security and Data Privacy

William Briguglio, Waleed A. Yousef, Waleed A. Yousef, Issa Traoré, Mohammad Mamun, Sherif Saad

2025

Abstract

Federated learning (FL) enables machine learning on data held across multiple clients without exchanging private data. However, exchanging information for model training can compromise data privacy. Further, participants may be untrustworthy and can attempt to sabotage model performance. Also, data that is not independently and identically distributed (IID) impede the convergence of FL techniques. We present a general framework for federated learning via aggregating multivariate estimated densities (FLAMED). FLAMED aggregates density estimations of clients’ data, from which it simulates training datasets to perform centralized learning, bypassing problems arising from non-IID data and contributing to addressing privacy and security concerns. FLAMED does not require a copy of the global model to be distributed to each participant during training, meaning the aggregating server can retain sole proprietorship of the global model without the use of resource-intensive homomorphic encryption. We compared its performance to standard FL approaches using synthetic and real datasets and evaluated its resilience to model poisoning attacks. Our results indicate that FLAMED effectively handles non-IID data in many settings while also being more secure.

Paper Citation

in Harvard Style

Briguglio W., Yousef W., Traoré I., Mamun M. and Saad S. (2025). An Alternative Approach to Federated Learning for Model Security and Data Privacy. In Proceedings of the 11th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP; ISBN 978-989-758-735-1, SciTePress, pages 291-301. DOI: 10.5220/0013237500003899

in Bibtex Style

@conference{icissp25,
author={William Briguglio and Waleed Yousef and Issa Traoré and Mohammad Mamun and Sherif Saad},
title={An Alternative Approach to Federated Learning for Model Security and Data Privacy},
booktitle={Proceedings of the 11th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP},
year={2025},
pages={291-301},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013237500003899},
isbn={978-989-758-735-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 11th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP
TI - An Alternative Approach to Federated Learning for Model Security and Data Privacy
SN - 978-989-758-735-1
AU - Briguglio W.
AU - Yousef W.
AU - Traoré I.
AU - Mamun M.
AU - Saad S.
PY - 2025
SP - 291
EP - 301
DO - 10.5220/0013237500003899
PB - SciTePress