# Adversarial training approach for local data debiasing

## TL;DR

This paper introduces GANsan, a novel adversarial training method that removes sensitive attributes from data to prevent discrimination, while maintaining data interpretability and utility, demonstrated through real dataset experiments.

## Contribution

The paper presents GANsan, a new local data debiasing approach using generative adversarial networks that preserves data interpretability and can be applied locally before data release.

## Key findings

- Effective removal of sensitive attributes demonstrated
- Trade-off between fairness and data utility shown
- Applicable to real-world datasets with promising results

## Abstract

The widespread use of automated decision processes in many areas of our society raises serious ethical issues concerning the fairness of the process and the possible resulting discriminations. In this work, we propose a novel approach called GANsan whose objective is to prevent the possibility of any discrimination i.e., direct and indirect) based on a sensitive attribute by removing the attribute itself as well as the existing correlations with the remaining attributes. Our sanitization algorithm GANsan is partially inspired by the powerful framework of generative adversarial networks (in particular the Cycle-GANs), which offers a flexible way to learn a distribution empirically or to translate between two different distributions. In contrast to prior work, one of the strengths of our approach is that the sanitization is performed in the same space as the original data by only modifying the other attributes as little as possible and thus preserving the interpretability of the sanitized data. As a consequence, once the sanitizer is trained, it can be applied to new data, such as for instance, locally by an individual on his profile before releasing it. Finally, experiments on a real dataset demonstrate the effectiveness of the proposed approach as well as the achievable trade-off between fairness and utility.