Author: Lydia T. Liu, Sarah Dean, Esther Rolf, Max Simchowitz, Moritz Hardt
Publisher: BAIR
Publication Year: 2018
Summary: The following article discusses the pursuit of fairness in terms of machine learning implementation in regards to treatment of disadvantaged groups that has led to a collection of intuitive “fairness criteria” (e.g., demographic parity, equal opportunity). The interactive article explains how blindly employing fairness criteria constraints without thought for the long-term well being outcomes for populations can work against intended efforts.