What-is and How-to for Fairness in Machine Learning: A Survey, Reflection, and Perspective
arXiv:2206.04101 · doi:10.1145/3597199
Abstract
Algorithmic fairness has attracted increasing attention in the machine learning community. Various definitions are proposed in the literature, but the differences and connections among them are not clearly addressed. In this paper, we review and reflect on various fairness notions previously proposed in machine learning literature, and make an attempt to draw connections to arguments in moral and political philosophy, especially theories of justice. We also consider fairness inquiries from a dynamic perspective, and further consider the long-term impact that is induced by current prediction and decision. In light of the differences in the characterized fairness, we present a flowchart that encompasses implicit assumptions and expected outcomes of different types of fairness inquiries on the data generating process, on the predicted outcome, and on the induced impact, respectively. This paper demonstrates the importance of matching the mission (which kind of fairness one would like to enforce) and the means (which spectrum of fairness analysis is of interest, what is the appropriate analyzing scheme) to fulfill the intended purpose.
References in corpus (19)
- Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment
- On the (im)possibility of fairness
- Roles for Computing in Social Change
- The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons
- Towards Long-term Fairness in Recommendation
- A Convex Framework for Fair Regression
- Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems
- Fairness in Algorithmic Decision Making: An Excursion Through the Lens of Causality
- Fairness in Risk Assessment Instruments: Post-Processing to Achieve Counterfactual Equalized Odds
- Calibrated Fairness in Bandits
- Survey on Causal-based Machine Learning Fairness Notions
- Achieving Equalized Odds by Resampling Sensitive Attributes
- The four-fifths rule is not disparate impact: a woeful tale of epistemic trespassing in algorithmic fairness
- Allocating Opportunities in a Dynamic Model of Intergenerational Mobility
- Principal Fairness for Human and Algorithmic Decision-Making
- Unintended Selection: Persistent Qualification Rate Disparities and Interventions
- Unfairness Despite Awareness: Group-Fair Classification with Strategic Agents
- Why Fair Labels Can Yield Unfair Predictions: Graphical Conditions for Introduced Unfairness
- Model Transferability With Responsive Decision Subjects