A Survey of Bias in Machine Learning Through the Prism of Statistical Parity.
Applications based on machine learning models have now become an indispensable part of the everyday life and the professional world. As a consequence, a critical question has recently arose among the population: Do algorithmic decisions convey any type of discrimination against specific groups of po...
| Publicado en: | American Statistician Vol. 76; no. 2; pp. 188 - 199 |
|---|---|
| Autores principales: | , , , , |
| Formato: | Artículo |
| Publicado: |
Taylor & Francis Ltd
May2022
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=156581445&site=ehost-live header: @attributes: shortDbName: ssf uiTerm: 156581445 longDbName: Social Sciences Full Text (H.W. Wilson) uiTag: AN controlInfo: bkinfo: jinfo: jid: 00031305 STT jtl: American Statistician issn: 00031305 maglogo: Y pubinfo: dt: May2022 vid: 76 iid: 2 pid: 377 pub: Taylor & Francis Ltd artinfo: ui: 156581445 10.1080/00031305.2021.1952897 ppf: 188 ppct: 11 formats: tig: atl: A Survey of Bias in Machine Learning Through the Prism of Statistical Parity. aug: au: Besse, Philippe del Barrio, Eustasio Gordaliza, Paula Loubes, Jean-Michel Risser, Laurent affil: Institut de Mathématiques de Toulouse, INSA, Université Toulouse 3, CNRS, Toulouse, France Instituto de Matemáticas de la Universidad de Valladolid, Dpto. de Estadistica e Investigacion Operativa, Universidad de Valladolid, Valladolid, Spain Basque Center for Applied Mathematics, Bilbao, Spain Artificial and Natural Intelligence Toulouse Institute (3IA ANITI), Toulouse, France su: Everyday life Machine learning sug: subj: Everyday life Machine learning keyword: Disparate impact Fairness Tutorial Disparate impact Fairness Tutorial ab: Applications based on machine learning models have now become an indispensable part of the everyday life and the professional world. As a consequence, a critical question has recently arose among the population: Do algorithmic decisions convey any type of discrimination against specific groups of population or minorities? In this article, we show the importance of understanding how bias can be introduced into automatic decisions. We first present a mathematical framework for the fair learning problem, specifically in the binary classification setting. We then propose to quantify the presence of bias by using the standard disparate impact index on the real and well-known adult income dataset. Finally, we check the performance of different approaches aiming to reduce the bias in binary classification outcomes. Importantly, we show that some intuitive methods are ineffective with respect to the statistical parity criterion. This sheds light on the fact that trying to make fair machine learning models may be a particularly challenging task, in particular when the training observations contain some bias. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: N holdings: @attributes: islocal: N |
|---|