XONN: XNOR-based Oblivious Deep Neural Network Inference

Authors: 

M. Sadegh Riazi and Mohammad Samragh, UC San Diego; Hao Chen, Kim Laine, and Kristin Lauter, Microsoft Research; Farinaz Koushanfar, UC San Diego

Abstract: 

Advancements in deep learning enable cloud servers to provide inference-as-a-service for clients. In this scenario, clients send their raw data to the server to run the deep learning model and send back the results. One standing challenge in this setting is to ensure the privacy of the clients' sensitive data. Oblivious inference is the task of running the neural network on the client's input without disclosing the input or the result to the server. This paper introduces XONN (pronounced /ZAn/), a novel end-to-end framework based on Yao's Garbled Circuits (GC) protocol, that provides a paradigm shift in the conceptual and practical realization of oblivious inference. In XONN, the costly matrix-multiplication operations of the deep learning model are replaced with XNOR operations that are essentially free in GC. We further provide a novel algorithm that customizes the neural network such that the runtime of the GC protocol is minimized without sacrificing the inference accuracy.

We design a user-friendly high-level API for XONN, allowing expression of the deep learning model architecture in an unprecedented level of abstraction. We further provide a compiler to translate the model description from high-level Python (i.e., Keras) to that of XONN. Extensive proof-of-concept evaluation on various neural network architectures demonstrates that XONN outperforms prior art such as Gazelle (USENIX Security'18) by up to 7×, MiniONN (ACM CCS'17) by 93×, and SecureML (IEEE S&P'17) by 37×. State-of-the-art frameworks require one round of interaction between the client and the server for each layer of the neural network, whereas, XONN requires a constant round of interactions for any number of layers in the model. XONN is first to perform oblivious inference on Fitnet architectures with up to 21 layers, suggesting a new level of scalability compared with state-of-the-art. Moreover, we evaluate XONN on four datasets to perform privacy-preserving medical diagnosis. The datasets include breast cancer, diabetes, liver disease, and Malaria.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {235489,
author = {M. Sadegh Riazi and Mohammad Samragh and Hao Chen and Kim Laine and Kristin Lauter and Farinaz Koushanfar},
title = {{XONN}: XNOR-based Oblivious Deep Neural Network Inference},
booktitle = {28th {USENIX} Security Symposium ({USENIX} Security 19)},
year = {2019},
isbn = {978-1-939133-04-5},
address = {Santa Clara, CA},
pages = {1501--1518},
url = {https://www.usenix.org/conference/usenixsecurity19/presentation/riazi},
publisher = {{USENIX} Association},
month = aug,
}