ON THE APPLICATION AND IMPACT OF ϵ-DP AND FAIRNESS IN AMBULANCE ENGAGEMENT TIME PREDICTION

Selene Cerna; Catuscia Palamidessi

ON THE APPLICATION AND IMPACT OF ϵ-DP AND FAIRNESS IN AMBULANCE ENGAGEMENT TIME PREDICTION

Selene Cerna
, Catuscia Palamidessi

Résultats de recherche: Contribution à une conférence › Papier › Revue par des pairs

Résumé

This study elaborates on a complete pipeline for the development of a private and fair Machine Learning (ML) model to predict ambulance engagement time. It was shown that sensitive variables reduced their impact on model building with Random Forest as the differential privacy budget (ϵ) decreased with the GRR and Geometric mechanisms. Also, the application of the Reweighing fairness mechanism negatively affected fairness in private models. Finally, it is possible to keep firefighters’ and victims’ privacy, recovering an ML model with good performance.

langue originale	Anglais
état	Publié - 1 janv. 2023
Evénement	1st Tiny Papers at 11th International Conference on Learning Representations, Tiny Papers @ ICLR 2023 - Kigali, Rwanda Durée: 5 mai 2023 → 5 mai 2023

Une conférence

Une conférence	1st Tiny Papers at 11th International Conference on Learning Representations, Tiny Papers @ ICLR 2023
Pays/Territoire	Rwanda
La ville	Kigali
période	5/05/23 → 5/05/23

Autres fichiers et liens

Lien vers la publication dans Scopus

Contient cette citation

@conference{c305aa79d1b44b9aadb4aad2480bda3d,

title = "ON THE APPLICATION AND IMPACT OF ϵ-DP AND FAIRNESS IN AMBULANCE ENGAGEMENT TIME PREDICTION",

abstract = "This study elaborates on a complete pipeline for the development of a private and fair Machine Learning (ML) model to predict ambulance engagement time. It was shown that sensitive variables reduced their impact on model building with Random Forest as the differential privacy budget (ϵ) decreased with the GRR and Geometric mechanisms. Also, the application of the Reweighing fairness mechanism negatively affected fairness in private models. Finally, it is possible to keep firefighters{\textquoteright} and victims{\textquoteright} privacy, recovering an ML model with good performance.",

author = "Selene Cerna and Catuscia Palamidessi",

note = "Publisher Copyright: {\textcopyright} 2023 1st Tiny Papers Track at ICLR 2023 - Tiny Papers @ ICLR 2023. All rights reserved.; 1st Tiny Papers at 11th International Conference on Learning Representations, Tiny Papers @ ICLR 2023 ; Conference date: 05-05-2023 Through 05-05-2023",

year = "2023",

month = jan,

day = "1",

language = "English",

}

ON THE APPLICATION AND IMPACT OF ϵ-DP AND FAIRNESS IN AMBULANCE ENGAGEMENT TIME PREDICTION. / Cerna, Selene; Palamidessi, Catuscia.
2023. Papier présenté � 1st Tiny Papers at 11th International Conference on Learning Representations, Tiny Papers @ ICLR 2023, Kigali, Rwanda.

Résultats de recherche: Contribution à une conférence › Papier › Revue par des pairs

TY - CONF

T1 - ON THE APPLICATION AND IMPACT OF ϵ-DP AND FAIRNESS IN AMBULANCE ENGAGEMENT TIME PREDICTION

AU - Cerna, Selene

AU - Palamidessi, Catuscia

PY - 2023/1/1

Y1 - 2023/1/1

N2 - This study elaborates on a complete pipeline for the development of a private and fair Machine Learning (ML) model to predict ambulance engagement time. It was shown that sensitive variables reduced their impact on model building with Random Forest as the differential privacy budget (ϵ) decreased with the GRR and Geometric mechanisms. Also, the application of the Reweighing fairness mechanism negatively affected fairness in private models. Finally, it is possible to keep firefighters’ and victims’ privacy, recovering an ML model with good performance.

AB - This study elaborates on a complete pipeline for the development of a private and fair Machine Learning (ML) model to predict ambulance engagement time. It was shown that sensitive variables reduced their impact on model building with Random Forest as the differential privacy budget (ϵ) decreased with the GRR and Geometric mechanisms. Also, the application of the Reweighing fairness mechanism negatively affected fairness in private models. Finally, it is possible to keep firefighters’ and victims’ privacy, recovering an ML model with good performance.

UR - https://www.scopus.com/pages/publications/105026756493

M3 - Paper

AN - SCOPUS:105026756493

T2 - 1st Tiny Papers at 11th International Conference on Learning Representations, Tiny Papers @ ICLR 2023

Y2 - 5 May 2023 through 5 May 2023

ER -

ON THE APPLICATION AND IMPACT OF ϵ-DP AND FAIRNESS IN AMBULANCE ENGAGEMENT TIME PREDICTION

Résumé

Une conférence

Autres fichiers et liens

Empreinte digitale

Contient cette citation