Objectives

The objective of this post is two-fold:

  1. Concept Explanations - I wish to explain the main concepts in fairness in very simple and practical terms. For the technically inclined, I would like to provide some depth using mathematics borrowed from Moritz Hardt et al., 2016

  2. Learning materials - Record here all the resources which I found useful while delving on the topic of fairness in machine learning

Concepts

  1. Sensitive attributes

  2. Demographic parity

  3. Equal odds

  4. Equal opportunity

  5. Fairness through Awareness

Resources

  1. Attacking discrimination in ML

  2. Equality of opportunity in ML

  3. Duke slides

  4. Fair ML book

  5. Moritz Hardt et al., Equality of opportunity in supervised learning