Published Papers and Related Projects

FHEON has been used in multiple research projects and publications focusing on privacy-preserving machine learning and homomorphic encryption.

FHEON Core Paper

FHEON: A configurable framework for privacy-preserving neural network inference using fully homomorphic encryption Nges Brian Njungle, Eric Jahns, and Michel A. Kinsy

This paper introduces FHEON, describing the architecture, HE-optimized CNN layers, and evaluation on multiple neural network models.

Reference:

@misc{njungle2025fheonconfigurableframeworkdeveloping,
  title={FHEON: A Configurable Framework for Developing Privacy-Preserving Neural Networks Using Homomorphic Encryption},
  author={Nges Brian Njungle and Eric Jahns and Michel A. Kinsy},
  year={2025},
  eprint={2510.03996},
  archivePrefix={arXiv},
  primaryClass={cs.CR},
  url={https://arxiv.org/abs/2510.03996}
}

Link: Read the paper

Highlighted Publications

Activate me!: Designing efficient activation functions for privacy-preserving machine learning with fully homomorphic encryption Nges Brian Njungle, Michel A. Kinsy

This examines different activation functions in privacy-preserving machine learning models using FHE. This functions include; the Square activation Function, ReLU Approximation and introduces a new approach for ReLU evaluation based on Scheme Switching.

Reference:

@inproceedings{njungle2025activate,
  title={Activate me!: Designing efficient activation functions for privacy-preserving machine learning with fully homomorphic encryption},
  author={Njungle, Nges Brian and Kinsy, Michel A},
  booktitle={International Conference on Cryptology in Africa},
  pages={51--73},
  year={2025},
  organization={Springer}
}

Link: Read the paper – Arxiv Version

Published Papers and Related Projects

FHEON Core Paper

Highlighted Publications

Other Related Projects