PLS-based approach for fair representation learning
Machine Learning
2025-02-25 v1 Computers and Society
Statistics Theory
Machine Learning
Statistics Theory
Abstract
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.
Cite
@article{arxiv.2502.16263,
title = {PLS-based approach for fair representation learning},
author = {Elena M. De-Diego and Adrián Perez-Suay and Paula Gordaliza and Jean-Michel Loubes},
journal= {arXiv preprint arXiv:2502.16263},
year = {2025}
}