Saved in:
Bibliographic Details
Main Authors: Liu, Kaixin, Xiong, Huixin, Duan, Bingyu, Cheng, Zexuan, Zhou, Xinyu, Zhang, Wanqian, Zhang, Xiangyu
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.04974
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929454119387136
author Liu, Kaixin
Xiong, Huixin
Duan, Bingyu
Cheng, Zexuan
Zhou, Xinyu
Zhang, Wanqian
Zhang, Xiangyu
author_facet Liu, Kaixin
Xiong, Huixin
Duan, Bingyu
Cheng, Zexuan
Zhou, Xinyu
Zhang, Wanqian
Zhang, Xiangyu
contents In the domain of cloud-based deep learning, the imperative for external computational resources coexists with acute privacy concerns, particularly identity leakage. To address this challenge, we introduce XNN and XNN-d, pioneering methodologies that infuse neural network features with randomized perturbations, striking a harmonious balance between utility and privacy. XNN, designed for the training phase, ingeniously blends random permutation with matrix multiplication techniques to obfuscate feature maps, effectively shielding private data from potential breaches without compromising training integrity. Concurrently, XNN-d, devised for the inference phase, employs adversarial training to integrate generative adversarial noise. This technique effectively counters black-box access attacks aimed at identity extraction, while a distilled face recognition network adeptly processes the perturbed features, ensuring accurate identification. Our evaluation demonstrates XNN's effectiveness, significantly outperforming existing methods in reducing identity leakage while maintaining a high model accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2408_04974
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle XNN: Paradigm Shift in Mitigating Identity Leakage within Cloud-Enabled Deep Learning
Liu, Kaixin
Xiong, Huixin
Duan, Bingyu
Cheng, Zexuan
Zhou, Xinyu
Zhang, Wanqian
Zhang, Xiangyu
Cryptography and Security
Computer Vision and Pattern Recognition
In the domain of cloud-based deep learning, the imperative for external computational resources coexists with acute privacy concerns, particularly identity leakage. To address this challenge, we introduce XNN and XNN-d, pioneering methodologies that infuse neural network features with randomized perturbations, striking a harmonious balance between utility and privacy. XNN, designed for the training phase, ingeniously blends random permutation with matrix multiplication techniques to obfuscate feature maps, effectively shielding private data from potential breaches without compromising training integrity. Concurrently, XNN-d, devised for the inference phase, employs adversarial training to integrate generative adversarial noise. This technique effectively counters black-box access attacks aimed at identity extraction, while a distilled face recognition network adeptly processes the perturbed features, ensuring accurate identification. Our evaluation demonstrates XNN's effectiveness, significantly outperforming existing methods in reducing identity leakage while maintaining a high model accuracy.
title XNN: Paradigm Shift in Mitigating Identity Leakage within Cloud-Enabled Deep Learning
topic Cryptography and Security
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.04974