FairRR: Pre-Processing for Group Fairness through Randomized Response
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917611729584128 |
|---|---|
| author | Zeng, Xianli Ward, Joshua Cheng, Guang |
| author_facet | Zeng, Xianli Ward, Joshua Cheng, Guang |
| contents | The increasing usage of machine learning models in consequential decision-making processes has spurred research into the fairness of these systems. While significant work has been done to study group fairness in the in-processing and post-processing setting, there has been little that theoretically connects these results to the pre-processing domain. This paper proposes that achieving group fairness in downstream models can be formulated as finding the optimal design matrix in which to modify a response variable in a Randomized Response framework. We show that measures of group fairness can be directly controlled for with optimal model utility, proposing a pre-processing algorithm called FairRR that yields excellent downstream model utility and fairness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_07780 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FairRR: Pre-Processing for Group Fairness through Randomized Response Zeng, Xianli Ward, Joshua Cheng, Guang Machine Learning The increasing usage of machine learning models in consequential decision-making processes has spurred research into the fairness of these systems. While significant work has been done to study group fairness in the in-processing and post-processing setting, there has been little that theoretically connects these results to the pre-processing domain. This paper proposes that achieving group fairness in downstream models can be formulated as finding the optimal design matrix in which to modify a response variable in a Randomized Response framework. We show that measures of group fairness can be directly controlled for with optimal model utility, proposing a pre-processing algorithm called FairRR that yields excellent downstream model utility and fairness. |
| title | FairRR: Pre-Processing for Group Fairness through Randomized Response |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2403.07780 |