Adjustable Privacy using Autoencoder-based Learning Structure

TL;DR

This paper introduces an autoencoder-based method that enables data providers to control privacy levels while maintaining data utility, improving privacy-utility trade-offs for image and categorical data.

Contribution

It presents a modified autoencoder structure that separates, anonymizes, and enhances features, allowing adjustable privacy settings and better performance than existing approaches.

Findings

Significant improvement in privacy-utility trade-off over previous methods

Effective separation of confidential and non-confidential features

Applicable to both image and categorical datasets

Abstract

Inference centers need more data to have a more comprehensive and beneficial learning model, and for this purpose, they need to collect data from data providers. On the other hand, data providers are cautious about delivering their datasets to inference centers in terms of privacy considerations. In this paper, by modifying the structure of the autoencoder, we present a method that manages the utility-privacy trade-off well. To be more precise, the data is first compressed using the encoder, then confidential and non-confidential features are separated and uncorrelated using the classifier. The confidential feature is appropriately combined with noise, and the non-confidential feature is enhanced, and at the end, data with the original data format is produced by the decoder. The proposed architecture also allows data providers to set the level of privacy required for confidential features. The proposed method has been examined for both image and categorical databases, and the results show a significant performance improvement compared to previous methods.