Abstract:Homomorphic Encryption (HE) is one of the most promising security solutions to emerging Machine Learning as a Service (MLaaS). Leveled-HE (LHE)-enabled Convolutional Neural Networks (LHECNNs) are proposed to implement MLaaS to avoid large bootstrapping overhead. However, prior LHECNNs have to pay significant computing overhead but achieve only low inference accuracy, due to their polynomial approximation activations and poolings. Stacking many polynomial approximation activation layers in a network greatly reduces inference accuracy, since the polynomial approximation activation errors lead to a low distortion of the output distribution of the next batch normalization layer. So the polynomial approximation activations and poolings have become the obstacle to a fast and accurate LHECNN model. In this paper, we propose a Shift-accumulation-based LHE-enabled deep neural network (SHE) for fast and accurate inferences on encrypted data. We use the binary-operation-friendly Leveled Fast Homomorphic Encryption over Torus (LTFHE) encryption scheme to implement ReLU activations and max poolings. We also adopt the logarithmic quantization to accelerate inferences by replacing expensive LTFHE multiplications with cheap LTFHE shifts. We propose a mixed bitwidth accumulator to accelerate accumulations. Since the LTFHE ReLU activations, max poolings, shifts and accumulations have small multiplicative depth overhead, SHE can implement much deeper network architectures with more convolutional and activation layers. Our experimental results show SHE achieves the state-of-the-art inference accuracy and reduces the inference latency by 76.21% ~ 94.23% over prior LHECNNs on MNIST and CIFAR-10. The source code of SHE is available at <a class="link-external link-https" href="https://github.com/qianlou/SHE" rel="external noopener nofollow">this https URL</a>.

Cryptotree: fast and accurate predictions on encrypted structured data

Efficient Encrypted Inference on Ensembles of Decision Trees

Privacy-Preserving Tree-Based Inference with TFHE

SoK: Modular and Efficient Private Decision Tree Evaluation.

A Hybrid-Domain Framework for Secure Gradient Tree Boosting.

Efficient and Privacy-Preserving Tree-Based Inference via Additive Homomorphic Encryption

Random forest evaluation using multi-key homomorphic encryption and lookup tables

Efficient CNN Building Blocks for Encrypted Data

EDLaaS: Fully Homomorphic Encryption Over Neural Network Graphs for Vision and Private Strawberry Yield Forecasting

XNET: A Real-Time Unified Secure Inference Framework Using Homomorphic Encryption.

MOFHEI: Model Optimizing Framework for Fast and Efficient Homomorphically Encrypted Neural Network Inference

Speed-up of Data Analysis with Kernel Trick in Encrypted Domain

Level Up: Private Non-Interactive Decision Tree Evaluation using Levelled Homomorphic Encryption

Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inference

Crypto-Nets: Neural Networks over Encrypted Data

TT-TFHE: a Torus Fully Homomorphic Encryption-Friendly Neural Network Architecture

Hyperdimensional Computing as a Rescue for Efficient Privacy-Preserving Machine Learning-as-a-Service

Efficient Clustering on Encrypted Data.

Privacy-preserving outsourcing decision tree evaluation from homomorphic encryption

SHE: A Fast and Accurate Deep Neural Network for Encrypted Data

Vectorized Secure Evaluation of Decision Forests