EUVIMEDEUVIMED entdecken
Crossref

Roulette: A Semantic Privacy-Preserving Device-Edge Collaborative Inference Framework for Deep Learning Classification Tasks

Jingyi Li, Guocheng Liao, Lin Chen, Xu Chen

IEEE Transactions on Mobile Computing · 2024-05 · 23 (5) · 5494-5510

DOI: 10.1109/tmc.2023.3312304 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2024-05. Verfasst von Jingyi Li und 3 weiteren Autor:innen. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 79 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Basic and Applied Basic Research Foundation of Guangdong Province, Pearl River Talent Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning classifiers are crucial in the age of artificial intelligence. The device-edge-based collaborative inference has been widely adopted as an efficient framework for promoting its applications in IoT and 5 G/6 G networks. However, it suffers from accuracy degradation under non-i.i.d. data distribution and privacy disclosure. For accuracy degradation, direct use of transfer learning and split learning is high cost and privacy issues remain. For privacy disclosure, cryptography-based approaches lead to a huge overhead. Other lightweight methods assume that the ground truth is non-sensitive and can be exposed. But for many applications, the ground truth is the user's crucial privacy-sensitive information. In this paper, we propose a framework of Roulette, which is a task-oriented semantic privacy-preserving collaborative inference framework for deep learning classifiers. More than input data, we treat the ground truth of the data as private information. We develop a novel paradigm of split learning where the back-end DNN is frozen and the front-end DNN is retrained to be both a feature extractor and an encryptor. Moreover, we provide a differential privacy guarantee and analyze the hardness of ground truth inference attacks. To validate the proposed Roulette, we conduct extensive performance evaluations using realistic datasets, which demonstrate that Roulette can effectively defend against various attacks and meanwhile achieve good model accuracy. In a situation where the non-i.i.d. is very severe, Roulette improves the inference accuracy by 21% averaged over benchmarks, while making the accuracy of discrimination attacks almost equivalent to random guessing.

Förderung

  • National Natural Science Foundation of China
  • Basic and Applied Basic Research Foundation of Guangdong Province
  • Pearl River Talent Program

Zitieren

APA

Li, J., Liao, G., Chen, L., & Chen, X. (2024). Roulette: A Semantic Privacy-Preserving Device-Edge Collaborative Inference Framework for Deep Learning Classification Tasks. IEEE Transactions on Mobile Computing, 23(5), 5494-5510. https://doi.org/10.1109/tmc.2023.3312304

Vancouver

Li J, Liao G, Chen L, Chen X. Roulette: A Semantic Privacy-Preserving Device-Edge Collaborative Inference Framework for Deep Learning Classification Tasks. IEEE Transactions on Mobile Computing. 2024;23(5):5494-5510. doi:10.1109/tmc.2023.3312304

BibTeX
@article{Li2024,
  author = {Li, Jingyi and Liao, Guocheng and Chen, Lin and Chen, Xu},
  title = {Roulette: A Semantic Privacy-Preserving Device-Edge Collaborative Inference Framework for Deep Learning Classification Tasks},
  journal = {IEEE Transactions on Mobile Computing},
  year = {2024},
  volume = {23},
  number = {5},
  pages = {5494--5510},
  doi = {10.1109/tmc.2023.3312304}
}
Kostenlos starten →