Boundary-Based Fairness Constraints in Decision Trees and Random Forests

Geraldin Nanfack; Valentin Delchevalerie; Benoît Frénay

doi:10.14428/esann/2021.ES2021-69

Boundary-Based Fairness Constraints in Decision Trees and Random Forests

Geraldin Nanfack, Valentin Delchevalerie, Benoît Frénay

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

105 Téléchargements (Pure)

Résumé

Decision Trees (DTs) and Random Forests (RFs) are popular models in Machine Learning (ML) thanks to their interpretability and efficiency to solve real-world problems. However, DTs may sometimes learn rules that treat different groups of people unfairly, by paying attention to sensitive features like for example gender, age, income, language, etc. Even if several solutions have been proposed to reduce the unfairness for different ML algorithms, few of them apply to DTs. This work aims to transpose a successful method proposed by Zafar et al. to reduce the unfairness in boundary based ML models to DTs.

langue originale	Anglais
titre	ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Pages	375-380
Nombre de pages	6
ISBN (Electronique)	9782875870827
Les DOIs	https://doi.org/10.14428/esann/2021.ES2021-69
Etat de la publication	Publié - 2021

Série de publications

Nom	ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

SDG des Nations Unies

Ce résultat contribue à ou aux Objectifs de développement durable suivants

Accès au document

10.14428/esann/2021.ES2021-69

ES2021-69Version finale publiée, 1,67 MB

Autres fichiers et liens

Lien vers la publication sur Scopus

Contient cette citation

Nanfack, G., Delchevalerie, V., & Frénay, B. (2021). Boundary-Based Fairness Constraints in Decision Trees and Random Forests. Dans ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (p. 375-380). (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning). https://doi.org/10.14428/esann/2021.ES2021-69

Nanfack, Geraldin ; Delchevalerie, Valentin ; Frénay, Benoît. / Boundary-Based Fairness Constraints in Decision Trees and Random Forests. ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. 2021. p. 375-380 (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning).

@inproceedings{3fb5150eb43f46a2b12f3be092688e8b,

title = "Boundary-Based Fairness Constraints in Decision Trees and Random Forests",

abstract = "Decision Trees (DTs) and Random Forests (RFs) are popular models in Machine Learning (ML) thanks to their interpretability and efficiency to solve real-world problems. However, DTs may sometimes learn rules that treat different groups of people unfairly, by paying attention to sensitive features like for example gender, age, income, language, etc. Even if several solutions have been proposed to reduce the unfairness for different ML algorithms, few of them apply to DTs. This work aims to transpose a successful method proposed by Zafar et al. [1] to reduce the unfairness in boundary based ML models to DTs.",

author = "Geraldin Nanfack and Valentin Delchevalerie and Beno{\^i}t Fr{\'e}nay",

note = "Funding Information: G.N. and V.D. were supported by the EOS VeriLearn project n. 30992574. V.D. is supported by the Walloon region with a Ph.D. grant from FRIA (F.R.S.-FNRS). This research used resources of PTCI at UNamur, supported by the F.R.S.-FNRS under the convention n. 2.5020.11. The authors thank Adrien Bibal for the fruitful discussions on this paper. Publisher Copyright: {\textcopyright} 2021 ESANN Intelligence and Machine Learning. All rights reserved.",

year = "2021",

doi = "10.14428/esann/2021.ES2021-69",

language = "English",

series = "ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

pages = "375--380",

booktitle = "ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

}

Nanfack, G , Delchevalerie, V & Frénay, B 2021, Boundary-Based Fairness Constraints in Decision Trees and Random Forests. Dans ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, p. 375-380. https://doi.org/10.14428/esann/2021.ES2021-69

Boundary-Based Fairness Constraints in Decision Trees and Random Forests. / Nanfack, Geraldin ; Delchevalerie, Valentin ; Frénay, Benoît.
ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. 2021. p. 375-380 (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning).

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

TY - GEN

T1 - Boundary-Based Fairness Constraints in Decision Trees and Random Forests

AU - Nanfack, Geraldin

AU - Delchevalerie, Valentin

AU - Frénay, Benoît

N1 - Funding Information: G.N. and V.D. were supported by the EOS VeriLearn project n. 30992574. V.D. is supported by the Walloon region with a Ph.D. grant from FRIA (F.R.S.-FNRS). This research used resources of PTCI at UNamur, supported by the F.R.S.-FNRS under the convention n. 2.5020.11. The authors thank Adrien Bibal for the fruitful discussions on this paper. Publisher Copyright: © 2021 ESANN Intelligence and Machine Learning. All rights reserved.

PY - 2021

Y1 - 2021

N2 - Decision Trees (DTs) and Random Forests (RFs) are popular models in Machine Learning (ML) thanks to their interpretability and efficiency to solve real-world problems. However, DTs may sometimes learn rules that treat different groups of people unfairly, by paying attention to sensitive features like for example gender, age, income, language, etc. Even if several solutions have been proposed to reduce the unfairness for different ML algorithms, few of them apply to DTs. This work aims to transpose a successful method proposed by Zafar et al. [1] to reduce the unfairness in boundary based ML models to DTs.

AB - Decision Trees (DTs) and Random Forests (RFs) are popular models in Machine Learning (ML) thanks to their interpretability and efficiency to solve real-world problems. However, DTs may sometimes learn rules that treat different groups of people unfairly, by paying attention to sensitive features like for example gender, age, income, language, etc. Even if several solutions have been proposed to reduce the unfairness for different ML algorithms, few of them apply to DTs. This work aims to transpose a successful method proposed by Zafar et al. [1] to reduce the unfairness in boundary based ML models to DTs.

UR - http://www.scopus.com/inward/record.url?scp=85129302481&partnerID=8YFLogxK

U2 - 10.14428/esann/2021.ES2021-69

DO - 10.14428/esann/2021.ES2021-69

M3 - Conference contribution

T3 - ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

SP - 375

EP - 380

BT - ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

ER -

Nanfack G , Delchevalerie V , Frénay B. Boundary-Based Fairness Constraints in Decision Trees and Random Forests. Dans ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. 2021. p. 375-380. (ESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning). doi: 10.14428/esann/2021.ES2021-69

Boundary-Based Fairness Constraints in Decision Trees and Random Forests

Résumé

Série de publications

SDG des Nations Unies

Accès au document

Autres fichiers et liens

Empreinte digitale

Contient cette citation