Data privacy and utility trade-off based on mutual information neural estimator

作者：

Highlights：

• The proposed framework can maximize data utility for multiple privacy budgets.

• The neural estimator could estimate mutual information with unknown distributions.

• Simulations show our method outperforms benchmark in privacy utility trade-off.

摘要

•The proposed framework can maximize data utility for multiple privacy budgets.•The neural estimator could estimate mutual information with unknown distributions.•Simulations show our method outperforms benchmark in privacy utility trade-off.

论文关键词：Privacy utility trade-off,Mutual information neural estimator,KL-divergence,Neural networks

论文评审过程：Received 10 May 2022, Revised 26 June 2022, Accepted 27 June 2022, Available online 3 July 2022, Version of Record 6 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118012