Privacy-Preserving in Machine Learning: Bridging Security and Performance in Modern Applications

Authors

  • Ramachandra Vamsi Krishna Nalam University of Buffalo, USA Author
  • Pooja Sri Nalam Seattle University, USA Author
  • Sruthi Anuvalasetty Seattle University, USA Author

DOI:

https://doi.org/10.32628/CSEIT25111262

Keywords:

Privacy-Preserving Machine Learning (PPML), Federated Learning, Homomorphic Encryption, Secure Multi-Party Computation (MPC), Data Privacy Architecture

Abstract

Privacy-preserving machine learning (PPML) has emerged as a critical paradigm in the era of data-driven applications, addressing the fundamental tension between leveraging large-scale datasets and protecting individual privacy. This technical article examines recent advances in PPML techniques, focusing on three key approaches: federated learning, which enables distributed model training while keeping data localized; homomorphic encryption, allowing computation on encrypted data; and secure multi-party computation (MPC) for privacy-conscious collaborative learning. Through detailed architectural analysis and real-world case studies in mobile device personalization and healthcare analytics, this article demonstrates how these techniques can be effectively implemented while navigating computational overhead and implementation complexity. This article reveals current PPML approaches successfully preserve privacy in production environments, but they face significant challenges in computational efficiency and system integration. This article concludes by presenting optimization strategies and emerging research directions aimed at making PPML more practical for large-scale deployments.

Downloads

Download data is not yet available.

References

Deval Parikh, Sarangkumar Radadia, "Privacy-Preserving Machine Learning Techniques: Challenges and Research Directions," ResearchGate, March 2024. [Online]. Available: https://www.researchgate.net/publication/379244515_Privacy-Preserving_Machine_Learning_Techniques_Challenges_And_Research_Directions

Alejandro Guerra-Manzanares et al., "Privacy-Preserving Machine Learning for Healthcare: Open Challenges and Future Perspectives," ICLR, 2023. [Online]. Available: https://openreview.net/pdf?id=4hsS1gZlPzW

Dhurgham Hassan Mahlool and Mohammed Hamzah Abed, "A Comprehensive Survey on Federated Learning: Concept and Applications," ResearchGate, January 2022. [Online]. Available: https://www.researchgate.net/publication/358148795_A_Comprehensive_Survey_on_Federated_Learning_Concept_and_Applications

Yonan Yonan, Mohammad Omar Abdullah, "Performance Evaluation of Privacy-Preserving Machine Learning for IoT," Halmstad University, 27 June 2024. [Online]. Available: https://hh.diva-portal.org/smash/get/diva2:1879256/FULLTEXT03.pdf

L Ohno-Machado et al., "Privacy, Security, and Machine Learning for Mobile Health Applications," American Association for the Advancement of Science (AAAS). [Online]. Available: https://www.aaas.org/sites/default/files/Ohno%20Privacy,%20Security,%20and%20Machine%20Learning%20for%20Mobile%20Health%20Applications%20.pdf

Nazish Khalid et al., "Privacy-preserving artificial intelligence in healthcare: Techniques and applications," ScienceDirect, Vol. 158, May 2023. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S001048252300313X

Runhua Xu et al., "Privacy-Preserving Machine Learning Methods: Challenges and Directions," ResearchGate, August 2021. [Online]. Available: https://www.researchgate.net/publication/353819224_Privacy-Preserving_Machine_Learning_Methods_Challenges_and_Directions

Payman Mohassel and Yupeng Zhang, "SecureML: A System for Scalable Privacy-Preserving Machine Learning," Cryptology ePrint Archieve, 2017. [Online]. Available: https://eprint.iacr.org/2017/396.pdf

James Joshi, "Privacy-Preserving Machine Learning Methods: Challenges and Directions," arXiv (Cornell University), 2021. [Online]. Available: https://www.academia.edu/99320431/Privacy_Preserving_Machine_Learning_Methods_Challenges_and_Directions

Dimpy Jindal et al., "Emerging trends of privacy and security in cloud computing," AIP Conference Proceedings, vol. 2760, no. 1, 2 June 2023. [Online]. Available: https://pubs.aip.org/aip/acp/article-abstract/2760/1/020013/2894021/Emerging-trends-of-privacy-and-security-in-cloud

Wagh, Sameer, "DataSpace: New Directions in Efficient Privacy-Preserving Machine Learning," Princeton University, Technical Report, 2020. [Online]. Available: https://dataspace.princeton.edu/handle/88435/dsp01s7526g34f

Dimpy Jindal et al., "Emerging trends of privacy and security in cloud computing," ResearchGate, January 2023. [Online]. Available: https://www.researchgate.net/publication/371268031_Emerging_trends_of_privacy_and_security_in_cloud_computing

Downloads

Published

20-01-2025

Issue

Section

Research Articles

How to Cite

Privacy-Preserving in Machine Learning: Bridging Security and Performance in Modern Applications. (2025). International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 11(1), 894-907. https://doi.org/10.32628/CSEIT25111262