Privacy Enhanced Federated Learning in Encrypted Anonymous Personal Device Domain

Abstract:

The increase in privacy concerns and the introduction of privacy and data protection legislation compel organisations to reevaluate their practices regarding traditional machine learning. The aggregation and management of users’ private data on the central server may contravene regulations if not properly administered. Federated learning provides a technique that eliminates the necessity of uploading users’ data to the server. It facilitates substantial learning by collaboratively training on each client’s devices and pooling the model gradient changes. Federated learning, augmented with a proxy as an intermediary and encrypted model parameters, will enhance anonymity, privacy, and data protection against malicious threats, including membership inference adversaries. Nonetheless, encrypted data incurs costs for customers’ communication and data size that exceed twice the original size. Our paper seeks to resolve these issues. We present two secure approaches for effective communication in an anonymous encrypted federated learning framework as our contribution. Additionally, our experiments demonstrated that it is feasible to attain equivalent communication costs as in non-encrypted scenarios. We provide recommendations in the conclusion for the effective implementation of privacy-preserving federated learning in the area of personal devices.

• Periodical: Engineering Headway (Volume 27)
• Pages: 3-12
• DOI: https://doi.org/10.4028/p-erHlI5
• Online since: October 2025
• Authors: Muhammad Senoyodha Brennaf*, Po Yang, Vitaveska Lanfranchi
• Keywords: Anonymous, Encryption, Federated Learning, Privacy, Proxy

References

[1] European Union, What is GDPR, the EU's new data protection law?, GDPR.eu, (2018). [Online]. Available: https://gdpr.eu/what-is-gdpr/.

DOI: 10.1093/law-oeeul/e137.013.137

[2] B. McMahan, E. Moore, D. Ramage, S. Hampson, B. A. y Arcas, Communication-efficient learning of deep networks from decentralized data, Artif. Intell. Statist. (2017) 1273–1282.

[3] P. Kairouz, et al., Advances and open problems in federated learning, Found. Trends® Mach. Learn. 14 (2021) 1–210.

[4] C. Dwork, A. Roth, The algorithmic foundations of differential privacy, Found. Trends Theor. Comput. Sci. 9 (2014) 211–407.

[5] K. Bonawitz, et al., Practical secure aggregation for privacy-preserving machine learning, Proc. 2017 ACM SIGSAC Conf. Comput. Commun. Secur. (2017) 1175–1191.

DOI: 10.1145/3133956.3133982

[6] C. Gentry, Fully homomorphic encryption using ideal lattices, Proc. 41st Annu. ACM Symp. Theory Comput. (2009) 169–178.

DOI: 10.1145/1536414.1536440

[7] R.C. Geyer, T. Klein, M. Nabi, Differentially private federated learning: A client level perspective, arXiv:1712.07557 (2017).

[8] S. Kadhe, N. Rajaraman, O. O. Koyluoglu, K. Ramchandran, Fastsecagg: Scalable secure aggregation for privacy-preserving federated learning, arXiv:2009.11248 (2020).

[9] C. Dilmegan, What is Homomorphic Encryption? Benefits & Challenges, AI Multiple (2022). [Online]. Available: https://research.aimultiple.com/homomorphic-encryption/.

[10] W. Wang, Y. Hu, L. Chen, X. Huang, B. Sunar, Exploring the feasibility of fully homomorphic encryption, IEEE Trans. Comput. 64 (2013) 698–706.

DOI: 10.1109/tc.2013.154

[11] J. Konečný, H. B. McMahan, F. X. Yu, P. Richtárik, A. T. Suresh, D. Bacon, Federated learning: Strategies for improving communication efficiency, arXiv:1610.05492 (2016).

[12] J. Konečný, H. B. McMahan, D. Ramage, P. Richtárik, Federated optimization: Distributed machine learning for on-device intelligence, arXiv:1610.02527 (2016).

[13] Q. Yang, Y. Liu, Y. Cheng, Y. Kang, T. Chen, H. Yu, Federated learning, Synth. Lect. Artif. Intell. Mach. Learn. 13 (2019) 1–207.

[14] E. Kinnear, P. McManus, T. Pauly, T. Verma, C. A. Wood, RFC 9230 - Oblivious DNS over HTTPS, IETF (2022). [Online]. Available: https://datatracker.ietf.org/doc/rfc9230/.

DOI: 10.17487/rfc9230

[15] G. Rosinosky, S. Da Silva, S. Ben Mokhtar, D. Négru, L. Réveillère, E. Rivière, PProx: Efficient privacy for recommendation-as-a-service, Proc. 22nd Int. Middleware Conf. (2021) 14–26.

DOI: 10.1145/3464298.3476130

[16] S. Kalra, J. Wen, J. C. Cresswell, M. Volkovs, H. R. Tizhoosh, ProxyFL: Decentralized Federated Learning through Proxy Model Sharing, arXiv:2111.11343 (2021).

DOI: 10.21203/rs.3.rs-1168002/v1

[17] H. Shi, V. Radu, P. Yang, Distributed Training for Speech Recognition using Local Knowledge Aggregation and Knowledge Distillation in Heterogeneous Systems, Proc. 3rd Workshop Mach. Learn. Syst. (2023) 64–70.

DOI: 10.1145/3578356.3592591

[18] M. Brennaf, P. Yang, V. Lanfranchi, A Comparative Analysis of Federated Learning Techniques on On-Demand Platforms in Supporting Modern Web Browser Applications, 2023 IEEE 22nd Int. Conf. Trust Secur. Priv. Comput. Commun. (TrustCom) (2023) 2601–2606.

DOI: 10.1109/trustcom60117.2023.00363

[19] EPFL, epfml/disco: Decentralized & federated privacy-preserving ML training, using p2p networking, in JS, (2022). [Online]. Available: https://github.com/epfml/disco.

[20] C. Deotte, 25 Million Images! [0.99757] MNIST Kaggle, Kaggle (2021). [Online]. Available: https://www.kaggle.com/code/cdeotte/25-million-images-0-99757-mnist.