Differentially Private Neural Networks with Bounded Activation Function

Kijung Jung, Hyukki Lee, Yon Dohn Chung

Research output: Contribution to journalArticlepeer-review

Abstract

Deep learning has shown outstanding performance in various fields, and it is increasingly deployed in privacy-critical domains. If sensitive data in the deep learning model are exposed, it can cause serious privacy threats. To protect individual privacy, we propose a novel activation function and stochastic gradient descent for applying differential privacy to deep learning. Through experiments, we show that the proposed method can effectively protect the privacy and the performance of proposed method is better than the previous approaches.

Original languageEnglish
Pages (from-to)905-908
Number of pages4
JournalIEICE Transactions on Information and Systems
Volume104
Issue number6
DOIs
Publication statusPublished - 2021

Keywords

  • Activation function
  • Deep learning
  • Differential privacy

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Differentially Private Neural Networks with Bounded Activation Function<sup>∗</sup>'. Together they form a unique fingerprint.

Cite this