Nowadays Machine Learning (ML) techniques are extensively adopted in many socially sensitive systems, thus requiring to carefully study the fairness of the decisions taken by such systems. Many approaches have been proposed to address and to make sure there is no bias against individuals or specific groups which might originally come from biased training datasets or algorithm design. In this regard, we propose a fairness enforcing approach called EiFFFeL --Enforcing Fairness in Forests by Flipping Leaves-- which exploits tree-based or leaf-based post-processing strategies to relabel leaves of selected decision trees of a given forest. Experimental results show that our approach achieves a user-defined group fairness degree without losing a significant amount of accuracy.

EiFFFeL: Enforcing Fairness in Forests by Flipping Leaves

Seyum Assefa Abebe
;
Claudio Lucchese;Salvatore Orlando
2022

Abstract

Nowadays Machine Learning (ML) techniques are extensively adopted in many socially sensitive systems, thus requiring to carefully study the fairness of the decisions taken by such systems. Many approaches have been proposed to address and to make sure there is no bias against individuals or specific groups which might originally come from biased training datasets or algorithm design. In this regard, we propose a fairness enforcing approach called EiFFFeL --Enforcing Fairness in Forests by Flipping Leaves-- which exploits tree-based or leaf-based post-processing strategies to relabel leaves of selected decision trees of a given forest. Experimental results show that our approach achieves a user-defined group fairness degree without losing a significant amount of accuracy.
Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing
File in questo prodotto:
File Dimensione Formato  
EiFFFeL.pdf

accesso aperto

Descrizione: EiFFFeL: Enforcing Fairness in Forests by Flipping Leaves
Tipologia: Documento in Pre-print
Licenza: Accesso gratuito (solo visione)
Dimensione 654.15 kB
Formato Adobe PDF
654.15 kB Adobe PDF Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/10278/3754554
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact