Providing a Philosophical Critique and Guidance of Fairness Metrics
It addresses the need for philosophical grounding in fairness metrics for computer scientists, but it is incremental as it primarily summarizes and critiques existing ideas without introducing new methods or data.
The paper tackles the challenge of integrating philosophical perspectives into fairness metrics in computer science, aiming to provide a crash course in fairness philosophy for computer scientists and roboticists, but it does not present specific results or concrete numbers.
In this project, I seek to present a summarization and unpacking of themes of fairness both in the field of computer science and philosophy. This is motivated by an increased dependence on notions of fairness in computer science and the millennia of thought on the subject in the field of philosophy. It is my hope that this acts as a crash course in $\textit{fairness philosophy}$ for the everyday computer scientist and specifically roboticist. This paper will consider current state-of-the-art ideas in computer science, specifically algorithmic fairness, as well as attempt to lay out a rough set of guidelines for metric fairness. Throughout the discussion of philosophy, we will return to a thought experiment posed by Cynthia Dwork on the question of randomness.