English

High-Resolution Convolutional Neural Networks on Homomorphically Encrypted Data via Sharding Ciphertexts

Cryptography and Security 2024-01-30 v2 Machine Learning

Abstract

Recently, Deep Convolutional Neural Networks (DCNNs) including the ResNet-20 architecture have been privately evaluated on encrypted, low-resolution data with the Residue-Number-System Cheon-Kim-Kim-Song (RNS-CKKS) homomorphic encryption scheme. We extend methods for evaluating DCNNs on images with larger dimensions and many channels, beyond what can be stored in single ciphertexts. Additionally, we simplify and improve the efficiency of the recently introduced multiplexed image format, demonstrating that homomorphic evaluation can work with standard, row-major matrix packing and results in encrypted inference time speedups by 4.66.5×4.6-6.5\times. We also show how existing DCNN models can be regularized during the training process to further improve efficiency and accuracy. These techniques are applied to homomorphically evaluate a DCNN with high accuracy on the high-resolution ImageNet dataset, achieving 80.2%80.2\% top-1 accuracy. We also achieve an accuracy of homomorphically evaluated CNNs on the CIFAR-10 dataset of 98.3%98.3\%.

Keywords

Cite

@article{arxiv.2306.09189,
  title  = {High-Resolution Convolutional Neural Networks on Homomorphically Encrypted Data via Sharding Ciphertexts},
  author = {Vivian Maloney and Richard F. Obrecht and Vikram Saraph and Prathibha Rama and Kate Tallaksen},
  journal= {arXiv preprint arXiv:2306.09189},
  year   = {2024}
}

Comments

14 pages, 9 figures

R2 v1 2026-06-28T11:06:03.549Z