MLAILGJul 1, 2018

Gradient Reversal Against Discrimination

arXiv:1807.00392v147 citations
AI Analysis

This addresses fairness issues in neural networks for applications requiring non-discriminatory predictions, but it appears incremental as it builds on existing fairness methods.

The authors tackled the problem of making arbitrary neural networks fair by introducing GRAD, a simplified method that improves both individual and group fairness with multi-attribute protection.

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.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes