Panther: Practical Secure Two-Party Neural Network Inference
Top Cited Papers
- 3 January 2025
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 1149-1162
Abstract
Secure two-party neural network (2P-NN) inference allows the server with a neural network model and the client with inputs to perform neural network inference without revealing their private data to each other. However, the state-of-the-art 2P-NN inference still suffers from large computation and communication overhead especially when used in ImageNet-scale deep neural networks. In this work, we design and build Panther, a lightweight and efficient secure 2P-NN inference system, which has great efficiency in evaluating 2P-NN inference while safeguarding the privacy of the server and the client. At the core of Panther, we have new protocols for 2P-NN inference. Firstly, we propose a customized homomorphic encryption scheme to reduce burdensome polynomial multiplications in the homomorphic encryption arithmetic circuit of linear protocols. Secondly, we present a more efficient and communication concise design for the millionaires’ protocol, which enables non-linear protocols with less communication cost. Our evaluations over three sought-after varying-scale deep neural networks show that Panther outperforms the state-of-the-art 2P-NN inference systems in terms of end-to-end runtime and communication overhead. Panther achieves state-of-the-art performance with up to 24.95× speedup for linear protocols and 6.40× speedup for non-linear protocols in WAN when compared to prior arts.Keywords
Funding Information
- National Key Research and Development Program of China (2024YFB3108700)
- National Natural Science Foundation of China (62372195, 62462054, 62166047)
- Natural Science Foundation Project of Jiangxi Province (20232BAB202007)
- CCF-Huawei Populus Grove Fund
This publication has 47 references indexed in Scilit:
- B-LNN: Inference-time linear model for secure neural network inferenceInformation Sciences, 2023
- Reinforcement learning architecture for cyber–physical–social AI: state-of-the-art and perspectivesArtificial Intelligence Review, 2023
- SecureBiNN: 3-Party Secure Computation for Binarized Neural Network Inference2022
- Silver: Silent VOLE and Oblivious Transfer from Hardness of Decoding Structured LDPC Codes2021
- TFHE: Fast Fully Homomorphic Encryption Over the TorusJournal of Cryptology, 2019
- Densely Connected Convolutional Networks2017
- Deep Residual Learning for Image Recognition2016
- On Ideal Lattices and Learning with Errors over Rings2010
- Extending Oblivious Transfers Efficiently2003
- Efficient Multiparty Protocols Using Circuit Randomization2001