Details
| Originalsprache | Englisch |
|---|---|
| Seiten (von - bis) | 1694-1702 |
| Seitenumfang | 9 |
| Fachzeitschrift | Proceedings of the VLDB Endowment |
| Jahrgang | 14 |
| Ausgabenummer | 9 |
| Frühes Online-Datum | 22 Okt. 2020 |
| Publikationsstatus | Veröffentlicht - Mai 2021 |
Abstract
One of the fundamental problems of machine ethics is to avoid the perpetuation and amplification of discrimination through machine learning applications. In particular, it is desired to exclude the influence of attributes with sensitive information, such as gender or race, and other causally related attributes on the machine learning task. The state-of-the-art bias reduction algorithm Capuchin breaks the causality chain of such attributes by adding and removing tuples. However, this horizontal approach can be considered invasive because it changes the data distribution. A vertical approach would be to prune sensitive features entirely. While this would ensure fairness without tampering with the data, it could also hurt the machine learning accuracy. Therefore, we propose a novel multi-objective feature selection strategy that leverages feature construction to generate more features that lead to both high accuracy and fairness. On three well-known datasets, our system achieves higher accuracy than other fairness-aware approaches while maintaining similar or higher fairness.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Informatik (sonstige)
- Informatik (insg.)
- Allgemeine Computerwissenschaft
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: Proceedings of the VLDB Endowment, Jahrgang 14, Nr. 9, 05.2021, S. 1694-1702.
Publikation: Beitrag in Fachzeitschrift › Konferenzaufsatz in Fachzeitschrift › Forschung › Peer-Review
}
TY - JOUR
T1 - Automated Feature Engineering for Algorithmic Fairness.
AU - Salazar, Ricardo
AU - Neutatz, Felix
AU - Abedjan, Ziawasch
N1 - Funding Information: The contribution of Felix Neutatz was funded by the German Ministry for Education and Research as BIFOLD - Berlin Institute for the Foundations of Learning and Data (ref. 01IS18025A and ref. 01IS18037A). We thank Babak Salimi for providing us with their source code.
PY - 2021/5
Y1 - 2021/5
N2 - One of the fundamental problems of machine ethics is to avoid the perpetuation and amplification of discrimination through machine learning applications. In particular, it is desired to exclude the influence of attributes with sensitive information, such as gender or race, and other causally related attributes on the machine learning task. The state-of-the-art bias reduction algorithm Capuchin breaks the causality chain of such attributes by adding and removing tuples. However, this horizontal approach can be considered invasive because it changes the data distribution. A vertical approach would be to prune sensitive features entirely. While this would ensure fairness without tampering with the data, it could also hurt the machine learning accuracy. Therefore, we propose a novel multi-objective feature selection strategy that leverages feature construction to generate more features that lead to both high accuracy and fairness. On three well-known datasets, our system achieves higher accuracy than other fairness-aware approaches while maintaining similar or higher fairness.
AB - One of the fundamental problems of machine ethics is to avoid the perpetuation and amplification of discrimination through machine learning applications. In particular, it is desired to exclude the influence of attributes with sensitive information, such as gender or race, and other causally related attributes on the machine learning task. The state-of-the-art bias reduction algorithm Capuchin breaks the causality chain of such attributes by adding and removing tuples. However, this horizontal approach can be considered invasive because it changes the data distribution. A vertical approach would be to prune sensitive features entirely. While this would ensure fairness without tampering with the data, it could also hurt the machine learning accuracy. Therefore, we propose a novel multi-objective feature selection strategy that leverages feature construction to generate more features that lead to both high accuracy and fairness. On three well-known datasets, our system achieves higher accuracy than other fairness-aware approaches while maintaining similar or higher fairness.
UR - http://www.scopus.com/inward/record.url?scp=85115160562&partnerID=8YFLogxK
U2 - 10.14778/3461535.3463474
DO - 10.14778/3461535.3463474
M3 - Conference article
VL - 14
SP - 1694
EP - 1702
JO - Proceedings of the VLDB Endowment
JF - Proceedings of the VLDB Endowment
IS - 9
ER -