In this paper we propose a practical human-in-the-loop approach for algorithmic fairness, utilizing the selective classification framework. We describe a classification model that abstains from making predictions in cases of unfairness or uncertainty. Any rejected predictions can be passed on to a human expert, to review the possible unfairness issues and make the decisions more just.
A Fair Selective Classifier to Put Humans in the Loop
Lenders D.;Pellungrini R.
;Giannotti Fosca;
2024
Abstract
In this paper we propose a practical human-in-the-loop approach for algorithmic fairness, utilizing the selective classification framework. We describe a classification model that abstains from making predictions in cases of unfairness or uncertainty. Any rejected predictions can be passed on to a human expert, to review the possible unfairness issues and make the decisions more just.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
paper_14.pdf
accesso aperto
Tipologia:
Published version
Licenza:
Creative Commons
Dimensione
1.15 MB
Formato
Adobe PDF
|
1.15 MB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
