Local Data Debiasing for Fairness Based on Generative Adversarial Training
Autor: | Rosin Claude Ngueveu, Ulrich Aïvodji, François Bidet, Alain Tapp, Sébastien Gambs |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2021 |
Předmět: |
lcsh:T55.4-60.8
Computer science Process (engineering) fairness 02 engineering and technology 010501 environmental sciences Space (commercial competition) Debiasing Machine learning computer.software_genre 01 natural sciences Training (civil) lcsh:QA75.5-76.95 Theoretical Computer Science Adversarial system 020204 information systems 0202 electrical engineering electronic engineering information engineering sanitization lcsh:Industrial engineering. Management engineering 0105 earth and related environmental sciences Interpretability Numerical Analysis business.industry generative adversarial network Contrast (statistics) Computational Mathematics Computational Theory and Mathematics Artificial intelligence lcsh:Electronic computers. Computer science business computer Generative grammar |
Zdroj: | Algorithms, Vol 14, Iss 87, p 87 (2021) Algorithms Volume 14 Issue 3 |
ISSN: | 1999-4893 |
Popis: | The widespread use of automated decision processes in many areas of our society raises serious ethical issues with respect to the fairness of the process and the possible resulting discrimination. To solve this issue, we propose a novel adversarial training approach called GANSan for learning a sanitizer 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 method GANSan is partially inspired by the powerful framework of generative adversarial networks (in particular 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, thus preserving the interpretability of the sanitized data. Consequently, once the sanitizer is trained, it can be applied to new data locally by an individual on their profile before releasing it. Finally, experiments on real datasets demonstrate the effectiveness of the approach as well as the achievable trade-off between fairness and utility. |
Databáze: | OpenAIRE |
Externí odkaz: |