A Fair Classifier Embracing Triplet Collapse

A. Martzloff; N. Posocco; Q. Ferré

A Fair Classifier Embracing Triplet Collapse

Machine Learning 2023-06-08 v1 Artificial Intelligence Computers and Society

Authors: A. Martzloff , N. Posocco , Q. Ferré

Abstract

In this paper, we study the behaviour of the triplet loss and show that it can be exploited to limit the biases created and perpetuated by machine learning models. Our fair classifier uses the collapse of the triplet loss when its margin is greater than the maximum distance between two points in the latent space, in the case of stochastic triplet selection.

Keywords

classification fairness in machine learning

Cite

@article{arxiv.2306.04400,
  title  = {A Fair Classifier Embracing Triplet Collapse},
  author = {A. Martzloff and N. Posocco and Q. Ferré},
  journal= {arXiv preprint arXiv:2306.04400},
  year   = {2023}
}

Comments

9 pages, 7 figures, CAp2023

Related papers

View all related →