ProxiMix: Enhancing Fairness with Proximity Samples in Subgroups
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917793115406336 |
|---|---|
| author | Hu, Jingyu Hong, Jun Du, Mengnan Liu, Weiru |
| author_facet | Hu, Jingyu Hong, Jun Du, Mengnan Liu, Weiru |
| contents | Many bias mitigation methods have been developed for addressing fairness issues in machine learning. We found that using linear mixup alone, a data augmentation technique, for bias mitigation, can still retain biases present in dataset labels. Research presented in this paper aims to address this issue by proposing a novel pre-processing strategy in which both an existing mixup method and our new bias mitigation algorithm can be utilized to improve the generation of labels of augmented samples, which are proximity aware. Specifically, we proposed ProxiMix which keeps both pairwise and proximity relationships for fairer data augmentation. We conducted thorough experiments with three datasets, three ML models, and different hyperparameters settings. Our experimental results showed the effectiveness of ProxiMix from both fairness of predictions and fairness of recourse perspectives. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_01145 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ProxiMix: Enhancing Fairness with Proximity Samples in Subgroups Hu, Jingyu Hong, Jun Du, Mengnan Liu, Weiru Machine Learning Artificial Intelligence Many bias mitigation methods have been developed for addressing fairness issues in machine learning. We found that using linear mixup alone, a data augmentation technique, for bias mitigation, can still retain biases present in dataset labels. Research presented in this paper aims to address this issue by proposing a novel pre-processing strategy in which both an existing mixup method and our new bias mitigation algorithm can be utilized to improve the generation of labels of augmented samples, which are proximity aware. Specifically, we proposed ProxiMix which keeps both pairwise and proximity relationships for fairer data augmentation. We conducted thorough experiments with three datasets, three ML models, and different hyperparameters settings. Our experimental results showed the effectiveness of ProxiMix from both fairness of predictions and fairness of recourse perspectives. |
| title | ProxiMix: Enhancing Fairness with Proximity Samples in Subgroups |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2410.01145 |