ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition
Authors:
- 1 June 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 9400-9409
Abstract
In this work, we propose ENSEI, a secure inference (SI) framework based on the frequency-domain secure convolution (FDSC) protocol for the efficient execution of image inference in the encrypted domain. Our observation is that, under the combination of homomorphic encryption and secret sharing, homomorphic convolution can be obliviously carried out in the frequency domain, significantly simplifying the related computations. We provide protocol designs and parameter derivations for number-theoretic transform (NTT) based FDSC. In the experiment, we thoroughly study the accuracy-efficiency trade-offs between time- and frequency-domain homomorphic convolution. With ENSEI, compared to the best known works, we achieve 5--11x online time reduction, up to 33x setup time reduction, and up to 10x reduction in the overall inference time. A further 33% of bandwidth reductions can be obtained on binary neural networks with only 3% of accuracy degradation on the CIFAR-10 dataset.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- Oblivious Neural Network Predictions via MiniONN Transformations2017
- SecureML: A System for Scalable Privacy-Preserving Machine Learning2017
- Membership Inference Attacks Against Machine Learning Models2017
- Optimal Functional-Unit Assignment for Heterogeneous Systems Under Timing ConstraintIEEE Transactions on Parallel and Distributed Systems, 2017
- Deep Residual Learning for Image Recognition2016
- (Leveled) Fully Homomorphic Encryption without BootstrappingACM Transactions on Computation Theory, 2014
- Ensure privacy and security in the process of medical image analysis2013
- Fully Homomorphic Encryption without Modulus Switching from Classical GapSVP2012
- Fully Homomorphic Encryption with Relatively Small Key and Ciphertext Sizes2010
- Protocols for secure computations1982