High-Resolution Convolutional Neural Networks on Homomorphically Encrypted Data via Sharding Ciphertexts
Year: 2023
Type: preprint
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 $4.6-6.5\times$. 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 $80.2\%$ top-1 accuracy. We also achieve an accuracy of homomorphically evaluated CNNs on the CIFAR-10 dataset of $98.3\%$.
Source: arXiv (Cornell University)
Institution: Johns Hopkins University Applied Physics Laboratory
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4380994396