Paper 2021/1688

Low-Complexity Deep Convolutional Neural Networks on Fully Homomorphic Encryption Using Multiplexed Parallel Convolutions

Eunsang Lee, Joon-Woo Lee, Junghyun Lee, Young-Sik Kim, Yongjune Kim, Jong-Seon No, and Woosuk Choi

Abstract

Recently, the standard ResNet-20 network was successfully implemented on residue number system variant Cheon-Kim-Kim-Song (RNS-CKKS) scheme using bootstrapping, but the implementation lacks practicality due to high latency and low security level. To improve the performance, we first minimize total bootstrapping runtime using multiplexed parallel convolution that collects sparse output data for multiple channels compactly. We also propose the \emph{imaginary-removing bootstrapping} to prevent the deep neural networks from catastrophic divergence during approximate ReLU operations. In addition, we optimize level consumptions and use lighter and tighter parameters. Simulation results show that we have 4.67$\times$ lower inference latency and 134$\times$ less amortized runtime (runtime per image) for ResNet-20 compared to the state-of-the-art previous work, and we achieve standard 128-bit security. Furthermore, we successfully implement ResNet-110 with high accuracy on the RNS-CKKS scheme for the first time.