Welcome to the resource topic for 2023/632
Title:
High-Throughput Deep Convolutional Neural Networks on Fully Homomorphic Encryption Using Channel-By-Channel Packing
Authors: Jung Hee Cheon, Minsik Kang, Taeseong Kim, Junyoung Jung, Yongdong Yeo
Abstract:Secure Machine Learning as a Service is a viable
solution where clients seek secure delegation of
the ML computation while protecting their sensi-
tive data. We propose an efficient method to se-
curely evaluate deep standard convolutional neu-
ral 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-multiple-
data capabilities in ciphertexts. Along with fur-
ther optimizations such as lazy rescaling, lazy
Baby-Step Giant-Step, and ciphertext level man-
agement, we could significantly reduce the com-
putational cost of standard ResNet inference. Sim-
ulation results show that our work has improve-
ments 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 pre-
vious 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.
ePrint: https://eprint.iacr.org/2023/632
See all topics related to this paper.
Feel free to post resources that are related to this paper below.
Example resources include: implementations, explanation materials, talks, slides, links to previous discussions on other websites.
For more information, see the rules for Resource Topics .