English

Gradient Reversal Against Discrimination

Machine Learning 2018-07-03 v1 Artificial Intelligence Machine Learning

Abstract

No methods currently exist for making arbitrary neural networks fair. In this work we introduce GRAD, a new and simplified method to producing fair neural networks that can be used for auto-encoding fair representations or directly with predictive networks. It is easy to implement and add to existing architectures, has only one (insensitive) hyper-parameter, and provides improved individual and group fairness. We use the flexibility of GRAD to demonstrate multi-attribute protection.

Keywords

Cite

@article{arxiv.1807.00392,
  title  = {Gradient Reversal Against Discrimination},
  author = {Edward Raff and Jared Sylvester},
  journal= {arXiv preprint arXiv:1807.00392},
  year   = {2018}
}

Comments

Proceedings of the 5'th Workshop on Fairness, Accountability and Transparency in Machine Learning, 2018

R2 v1 2026-06-23T02:47:30.096Z