arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.05905cs.CRcs.LG
收藏

ShadowNet:一种面向卷积神经网络的安全高效端侧模型推理系统

ShadowNet: A Secure and Efficient On-device Model Inference System for Convolutional Neural Networks

Google(谷歌);Florida International University(佛罗里达国际大学);Northeastern University(东北大学);University of California, San Diego(加州大学圣地亚哥分校);University of Wisconsin-Madison(威斯康星大学麦迪逊分校)

Zhichuang Sun, Ruimin Sun, Changming Liu, Amrita Roy Chowdhury, Long Lu, Somesh Jha

AI总结:

ShadowNet利用TEE保护端侧CNN模型隐私,通过权重变换将线性层安全外包给不可信AI加速器,并在TEE内恢复结果,在五个流行CNN上验证了其安全性与实用性。

摘要:

随着移动和边缘设备上AI加速器使用量的增加,端侧机器学习(ML)正日益普及。如今,成千上万的专有ML模型被部署在数十亿台不可信设备上,这引发了关于模型隐私的严重安全担忧。然而,在不失去对不可信AI加速器访问能力的前提下保护模型隐私是一个具有挑战性的问题。本文提出了一种新颖的端侧模型推理系统ShadowNet。ShadowNet利用可信执行环境(TEE)保护模型隐私,同时将模型中的重型线性层安全地外包给不可信的硬件加速器。ShadowNet通过在外包前对线性层权重进行变换,并在TEE内部恢复结果来实现这一目标。非线性层同样在TEE内部保持安全。ShadowNet的设计确保了权重的高效变换以及后续结果的恢复。我们基于TensorFlow Lite构建了ShadowNet原型,并在五个流行的CNN上进行了评估,即MobileNet、ResNet-44、MiniVGG、ResNet-404和YOLOv4-tiny。评估结果表明,ShadowNet在提供强大安全保障的同时保持了合理的性能,为安全的端侧模型推理提供了一种实用解决方案。

英文摘要:

With the increased usage of AI accelerators on mobile and edge devices, on-device machine learning (ML) is gaining popularity. Thousands of proprietary ML models are being deployed today on billions of untrusted devices. This raises serious security concerns about model privacy. However, protecting model privacy without losing access to the untrusted AI accelerators is a challenging problem. In this paper, we present a novel on-device model inference system, ShadowNet. ShadowNet protects the model privacy with Trusted Execution Environment (TEE) while securely outsourcing the heavy linear layers of the model to the untrusted hardware accelerators. ShadowNet achieves this by transforming the weights of the linear layers before outsourcing them and restoring the results inside the TEE. The non-linear layers are also kept secure inside the TEE. ShadowNet's design ensures efficient transformation of the weights and the subsequent restoration of the results. We build a ShadowNet prototype based on TensorFlow Lite and evaluate it on five popular CNNs, namely, MobileNet, ResNet-44, MiniVGG, ResNet-404, and YOLOv4-tiny. Our evaluation shows that ShadowNet achieves strong security guarantees with reasonable performance, offering a practical solution for secure on-device model inference.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑