Adversarial Learned Fair Representations using Dampening and Stacking

Max Knobbout

Adversarial Learned Fair Representations using Dampening and Stacking

Machine Learning 2022-03-17 v1 Artificial Intelligence Computers and Society

Authors: Max Knobbout

Abstract

As more decisions in our daily life become automated, the need to have machine learning algorithms that make fair decisions increases. In fair representation learning we are tasked with finding a suitable representation of the data in which a sensitive variable is censored. Recent work aims to learn fair representations through adversarial learning. This paper builds upon this work by introducing a novel algorithm which uses dampening and stacking to learn adversarial fair representations. Results show that that our algorithm improves upon earlier work in both censoring and reconstruction.

Keywords

representation learning adversarial examples fairness in machine learning

Cite

@article{arxiv.2203.08637,
  title  = {Adversarial Learned Fair Representations using Dampening and Stacking},
  author = {Max Knobbout},
  journal= {arXiv preprint arXiv:2203.08637},
  year   = {2022}
}

Comments

8 pages, 3 figures

Related papers

View all related →