Fairness and machine learning
Limitations and Opportunities
This online textbook is an incomplete work in progress. Essential chapters are still missing. In the spirit of open review, we solicit broad feedback that will influence existing chapters, as well as the development of later material.
Table of contents
We introduce formal non-discrimination criteria, establish their relationships, and illustrate their limitations.
- Legal background and normative questions
We survey the literature on discrimination in law, sociology, and philosophy. We then discuss the challenges that arise in translating these ideas of fairness to the statistical decision-making setting.
We dive into the rich technical repertoire of causal inference and how it helps articulate and address shortcomings of the classification paradigm, while raising new conceptual and normative questions.
- Measurement
Considerations of fairness begin with measurement, the process of assigning empirical relationships to numerical representations.
- Algorithmic interventions
We survey and systematize a burgeoning set of algorithmic interventions aimed at promoting fairness, while highlighting the limitations of this paradigm.
- Related concepts
We discuss privacy, interpretability, transparency, and accountability, and how each of these concepts relates to fairness.
- Appendix: Technical background
PDF version (all available chapters)
Video tutorials
Course materials
Contact us
We welcome your feedback. For all book related questions and comments, please send an email to contact@fairmlbook.org.
Citations
To cite this book, please use this bibtex entry:
@book{barocas-hardt-narayanan,
title = {Fairness and Machine Learning},
author = {Solon Barocas and Moritz Hardt and Arvind Narayanan},
publisher = {fairmlbook.org},
note = {\url{http://www.fairmlbook.org}},
year = {2019}
}