PLS-based approach for fair representation learning
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916626034589696 |
|---|---|
| author | De-Diego, Elena M. Perez-Suay, Adrián Gordaliza, Paula Loubes, Jean-Michel |
| author_facet | De-Diego, Elena M. Perez-Suay, Adrián Gordaliza, Paula Loubes, Jean-Michel |
| contents | We revisit the problem of fair representation learning by proposing Fair Partial Least Squares (PLS) components. PLS is widely used in statistics to efficiently reduce the dimension of the data by providing representation tailored for the prediction. We propose a novel method to incorporate fairness constraints in the construction of PLS components. This new algorithm provides a feasible way to construct such features both in the linear and the non linear case using kernel embeddings. The efficiency of our method is evaluated on different datasets, and we prove its superiority with respect to standard fair PCA method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_16263 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PLS-based approach for fair representation learning De-Diego, Elena M. Perez-Suay, Adrián Gordaliza, Paula Loubes, Jean-Michel Machine Learning Computers and Society Statistics Theory We revisit the problem of fair representation learning by proposing Fair Partial Least Squares (PLS) components. PLS is widely used in statistics to efficiently reduce the dimension of the data by providing representation tailored for the prediction. We propose a novel method to incorporate fairness constraints in the construction of PLS components. This new algorithm provides a feasible way to construct such features both in the linear and the non linear case using kernel embeddings. The efficiency of our method is evaluated on different datasets, and we prove its superiority with respect to standard fair PCA method. |
| title | PLS-based approach for fair representation learning |
| topic | Machine Learning Computers and Society Statistics Theory |
| url | https://arxiv.org/abs/2502.16263 |