Legal requirements on explainability in machine learning

Adrien Bibal; Michael Lognoul; Alexandre De Streel; Benoît Frénay

doi:10.1007/s10506-020-09270-4

Legal requirements on explainability in machine learning

Adrien Bibal, Michael Lognoul, Alexandre De Streel, Benoît Frénay

Résultats de recherche: Contribution à un journal/une revue › Article › Revue par des pairs

362 Téléchargements (Pure)

Résumé

Deep learning and other black-box models are becoming more and more popular today. Despite their high performance, they may not be accepted ethically or legally because of their lack of explainability. This paper presents the increasing number of legal requirements on machine learning model interpretability and explainability in the context of private and public decision making. It then explains how those legal requirements can be implemented into machine-learning models and concludes with a call for more inter-disciplinary research on explainability.

langue originale	Anglais
Pages (de - à)	149-169
Nombre de pages	21
journal	Artificial Intelligence & Law
Volume	29
Numéro de publication	2
Les DOIs	https://doi.org/10.1007/s10506-020-09270-4 https://doi.org/10.1007/s10506-020-09270-4
Etat de la publication	Publié - juin 2021

Accès au document

8613Version finale publiée, 849 KB

Autres fichiers et liens

Lien vers la publication sur Scopus

Interpretability and Explainability in Machine Learning and their Application to Nonlinear Dimensionality Reduction
Auteur: Bibal, A., 16 nov. 2020
Superviseur: FRENAY, B. (Promoteur), VANHOOF, W. (Président), Cleve, A. (Jury), Dumas, B. (Jury), Lee, J. A. (Personne externe) (Jury) & Galarraga, L. A. (Personne externe) (Jury)
Student thesis: Doc types › Docteur en Sciences

Fichier

Contient cette citation

@article{a4f5698841d84f7b922947abe7f51a2c,

title = "Legal requirements on explainability in machine learning",

abstract = "Deep learning and other black-box models are becoming more and more popular today. Despite their high performance, they may not be accepted ethically or legally because of their lack of explainability. This paper presents the increasing number of legal requirements on machine learning model interpretability and explainability in the context of private and public decision making. It then explains how those legal requirements can be implemented into machine-learning models and concludes with a call for more inter-disciplinary research on explainability.",

keywords = "Explainability, Interpretability, Law, Machine learning",

author = "Adrien Bibal and Michael Lognoul and {De Streel}, Alexandre and Beno{\^i}t Fr{\'e}nay",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature B.V.",

year = "2021",

month = jun,

doi = "10.1007/s10506-020-09270-4",

language = "English",

volume = "29",

pages = "149--169",

journal = "Artificial Intelligence & Law",

issn = "1572-8382",

publisher = "Springer",

number = "2",

}

TY - JOUR

T1 - Legal requirements on explainability in machine learning

AU - Bibal, Adrien

AU - Lognoul, Michael

AU - De Streel, Alexandre

AU - Frénay, Benoît

PY - 2021/6

Y1 - 2021/6

N2 - Deep learning and other black-box models are becoming more and more popular today. Despite their high performance, they may not be accepted ethically or legally because of their lack of explainability. This paper presents the increasing number of legal requirements on machine learning model interpretability and explainability in the context of private and public decision making. It then explains how those legal requirements can be implemented into machine-learning models and concludes with a call for more inter-disciplinary research on explainability.

AB - Deep learning and other black-box models are becoming more and more popular today. Despite their high performance, they may not be accepted ethically or legally because of their lack of explainability. This paper presents the increasing number of legal requirements on machine learning model interpretability and explainability in the context of private and public decision making. It then explains how those legal requirements can be implemented into machine-learning models and concludes with a call for more inter-disciplinary research on explainability.

KW - Explainability

KW - Interpretability

KW - Law

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85088816026&partnerID=8YFLogxK

U2 - 10.1007/s10506-020-09270-4

DO - 10.1007/s10506-020-09270-4

M3 - Article

AN - SCOPUS:85088816026

SN - 1572-8382

VL - 29

SP - 149

EP - 169

JO - Artificial Intelligence & Law

JF - Artificial Intelligence & Law

IS - 2

ER -

Legal requirements on explainability in machine learning

Résumé

Accès au document

Autres fichiers et liens

Empreinte digitale

Thèses de l'étudiant

Interpretability and Explainability in Machine Learning and their Application to Nonlinear Dimensionality Reduction

Contient cette citation