Popcorn: Paillier Meets Compression For Efficient Oblivious Neural Network Inference
Abstract
Oblivious inference enables the cloud to provide neural network inference-as-a-service (NN-IaaS), whilst neither disclosing the client data nor revealing the server's model. However, the privacy guarantee under oblivious inference usually comes with a heavy cost of efficiency and accuracy. We propose Popcorn, a concise oblivious inference framework entirely built on the Paillier homomorphic encryption scheme. We design a suite of novel protocols to compute non-linear activation and max-pooling layers. We leverage neural network compression techniques (i.e., neural weights pruning and quantization) to accelerate the inference computation. To implement the Popcorn framework, we only need to replace algebraic operations of existing networks with their corresponding Paillier homomorphic operations, which is extremely friendly for engineering development. We first conduct the performance evaluation and comparison based on the MNIST and CIFAR-10 classification tasks. Compared with existing solutions, Popcorn brings a significant communication overhead deduction, with a moderate runtime increase. Then, we benchmark the performance of oblivious inference on ImageNet. To our best knowledge, this is the first report based on a commercial-level dataset, taking a step towards the deployment to production.
Keywords
Cite
@article{arxiv.2107.01786,
title = {Popcorn: Paillier Meets Compression For Efficient Oblivious Neural Network Inference},
author = {Jun Wang and Chao Jin and Souhail Meftah and Khin Mi Mi Aung},
journal= {arXiv preprint arXiv:2107.01786},
year = {2021}
}
Comments
This version corrects a naive but significant typo in Table 8 in our previous version. Previously, we mistakenly indicated the communication cost in COM(g), i.e., communication bandwidth in gigabyte. In fact, it should be COM as in this version, and COM presents the communication bandwidth in megabyte. It is megabyte, not gigabyte