Loading...

Deep Private-feature extraction

Osia, S. A ; Sharif University of Technology | 2020

651 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/TKDE.2018.2878698
  3. Publisher: IEEE Computer Society , 2020
  4. Abstract:
  5. We present and evaluate Deep Private-Feature Extractor (DPFE), a deep model which is trained and evaluated based on information theoretic constraints. Using the selective exchange of information between a user's device and a service provider, DPFE enables the user to prevent certain sensitive information from being shared with a service provider, while allowing them to extract approved information using their model. We introduce and utilize the log-rank privacy, a novel measure to assess the effectiveness of DPFE in removing sensitive information and compare different models based on their accuracy-privacy trade-off. We then implement and evaluate the performance of DPFE on smartphones to understand its complexity, resource demands, and efficiency trade-offs. Our results on benchmark image datasets demonstrate that under moderate resource utilization, DPFE can achieve high accuracy for primary tasks while preserving the privacy of sensitive information. © 1989-2012 IEEE
  6. Keywords:
  7. Privacy ; Data privacy ; Data structures ; Deep learning ; Economic and social effects ; Extraction ; Feature extraction ; Information theory ; Job analysis ; Personnel training ; Exchange of information ; Feature extractor ; Image datasets ; Resource demands ; Resource utilizations ; Sensitive informations ; Service provider ; Task analysis ; Information use
  8. Source: IEEE Transactions on Knowledge and Data Engineering ; Volume 32, Issue 1 , 2020 , Pages 54-66
  9. URL: https://ieeexplore.ieee.org/document/8515092