Skip to main navigation Skip to search Skip to main content

Differentially private convolutional neural networks with adaptive gradient descent

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Deep learning achieves remarkable success in the fields of target detection, computer vision, natural language processing, and speech recognition. However, traditional deep learning models may suffer the privacy risk due to some training data involve sensitive information, such as the medical histories, location information and face images. Attackers can exploit the implicit information to recover the sensitive information from the training data. In order to protecting privacy of deep learning model, we develop a novel optimization algorithm called DPAGD-CNN for convolution neural network which cooperates differential privacy technique. Specifically, DPAGD-CNN allocates privacy budgets more carefully in each iteration, rather than assigning a fixed privacy budget per iteration. We theoretically prove that our approach can protect the privacy of training data and it achieves higher classification accuracy under the moderate privacy budget in the MNIST and CIFAR-10 datasets.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE 4th International Conference on Data Science in Cyberspace, DSC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages642-648
Number of pages7
ISBN (Electronic)9781728145280
DOIs
StatePublished - Jun 2019
Externally publishedYes
Event4th IEEE International Conference on Data Science in Cyberspace, DSC 2019 - Hangzhou, China
Duration: 23 Jun 201925 Jun 2019

Publication series

NameProceedings - 2019 IEEE 4th International Conference on Data Science in Cyberspace, DSC 2019
Volume2019-January

Conference

Conference4th IEEE International Conference on Data Science in Cyberspace, DSC 2019
Country/TerritoryChina
CityHangzhou
Period23/06/1925/06/19

Keywords

  • Convolutional neural network
  • Differential privacy
  • Gradient descent
  • Privacy-preserving

Fingerprint

Dive into the research topics of 'Differentially private convolutional neural networks with adaptive gradient descent'. Together they form a unique fingerprint.

Cite this