Paper 2023/632

High-Throughput Deep Convolutional Neural Networks on Fully Homomorphic Encryption Using Channel-By-Channel Packing

Jung Hee Cheon, Seoul National University, CryptoLab. Inc.
Minsik Kang, Seoul National University
Taeseong Kim, Seoul National University
Junyoung Jung, Seoul National University
Yongdong Yeo, Seoul National University

Secure Machine Learning as a Service is a viable solution where clients seek secure delegation of the ML computation while protecting their sensitive data. We propose an efficient method to securely evaluate deep standard convolutional neural networks based on CKKS fully homomorphic encryption, in the manner of batch inference. In this paper, we introduce a packing method called Channel-by-Channel Packing that maximizes the slot compactness and single-instruction-multipledata capabilities in ciphertexts. Along with further optimizations such as lazy rescaling, lazy Baby-Step Giant-Step, and ciphertext level management, we could significantly reduce the computational cost of standard ResNet inference. Simulation results show that our work has improvements in amortized time by 5.04× (from 79.46s to 15.76s) and 5.20×(from 455.56s to 87.60s) for ResNet-20 and ResNet-110, compared to the previous best results, resp. We also got a dramatic reduction in memory usage for rotation keys from several hundred GBs to 6.91GB, which is about 38× smaller than the previous result.

Available format(s)
Publication info
Privacy-Preserving Machine LearningFully Homomorphic EncryptionConvoluional Neual Network (CNN)ResNet
Contact author(s)
jhcheon @ snu ac kr
kaiser351 @ snu ac kr
kts1023 @ snu ac kr
jhaeg0312 @ snu ac kr
yongdong @ snu ac kr
2023-05-04: last of 2 revisions
2023-05-03: received
See all versions
Short URL
Creative Commons Attribution


      author = {Jung Hee Cheon and Minsik Kang and Taeseong Kim and Junyoung Jung and Yongdong Yeo},
      title = {High-Throughput Deep Convolutional Neural Networks on Fully Homomorphic Encryption Using Channel-By-Channel Packing},
      howpublished = {Cryptology ePrint Archive, Paper 2023/632},
      year = {2023},
      note = {\url{}},
      url = {}
Note: In order to protect the privacy of readers, does not use cookies or embedded third party content.