Privacy-Preserving Feature Engineering for Federated Learning Analytics

  • Authors

    • Dr. Yuki Nakamura Associate Professor, Kyoto University, Japan Author
    • Olivia Martin Product Manager, Microsoft, USA Author

    DOI:

    https://doi.org/10.67228/30715717/IJDEIC-2021PI7Q1B

    Published 05-13-2021

  • Federated Learning (FL), Privacy-Preserving Machine Learning, Feature Engineering, Differential Privacy, Homomorphic Encryption, Secure Multi-Party Computation (SMPC), Vertical and Horizontal Federated Learning, Data Confidentiality, Decentralized Analytics, Secure Feature Transformation

    Issue

    Vol. 4 No. 1 (2021)

    Section

    Articles

    How to Cite

    [1]
    Y. Nakamura and O. Martin, “Privacy-Preserving Feature Engineering for Federated Learning Analytics”, IJDEIC, vol. 4, no. 1, pp. 01–09, May 2021, doi: 10.67228/30715717/IJDEIC-2021PI7Q1B.

  • Abstract

    Federated learning (FL) offers a decentralized approach to machine learning that preserves data privacy by training models locally across distributed devices. However, the feature engineering process—an essential step in improving model performance—often remains centralized or privacy-invasive, risking sensitive data exposure. This paper proposes a comprehensive framework for privacy-preserving feature engineering (PPFE) within federated learning analytics. We explore techniques such as homomorphic encryption, differential privacy, and secure multi-party computation to enable robust, privacy-safe feature selection, transformation, and extraction across clients. Our framework includes both vertical and horizontal FL settings and evaluates the trade-offs between privacy, utility, and communication overhead. Experimental results on real-world datasets demonstrate that our PPFE methods can significantly improve model performance without compromising data privacy. This work contributes towards building a more secure and efficient FL pipeline that ensures end-to-end data confidentiality.

  • References

    [1] McMahan, H. B., et al. (2017). Communication-efficient learning of deep networks from decentralized data. AISTATS.

    [2] Bonawitz, K., et al. (2017). Practical secure aggregation for privacy-preserving machine learning. CCS.

    [3] Dwork, C., & Roth, A. (2014). The algorithmic foundations of differential privacy. Foundations and Trends in Theoretical Computer Science.

    [4] Gentry, C. (2009). A fully homomorphic encryption scheme. PhD thesis, Stanford University.

    [5] Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020). Federated learning: Challenges, methods, and future directions. IEEE Signal Processing Magazine.

    [6] Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019). Federated machine learning: Concept and applications. ACM Transactions on Intelligent Systems and Technology.

    [7] Mohassel, P., & Zhang, Y. (2017). SecureML: A system for scalable privacy-preserving machine learning. IEEE S&P.

    [8] Hard, A., et al. (2018). Federated learning for mobile keyboard prediction. arXiv preprint arXiv:1811.03604.

    [9] Li, X., et al. (2021). Differentially private feature selection via stability maximization. ICML.

    [10] Zhang, T., et al. (2020). BatchCrypt: Efficient homomorphic encryption for cross-silo federated learning. USENIX Security Symposium.

    [11] Shokri, R., & Shmatikov, V. (2015). Privacy-preserving deep learning. CCS.

    [12] Xu, J., et al. (2021). FedCor: Feature correlation-guided federated learning. NeurIPS.

    [13] Melis, L., et al. (2019). Exploiting unintended feature leakage in collaborative learning. IEEE S&P.

    [14] Xu, Y., et al. (2019). A hybrid federated learning framework for privacy-preserving feature engineering. arXiv preprint arXiv:1905.06655.

  • Downloads