Developing a Philosophical Framework for Fair Machine Learning: The Case of Algorithmic Collusion and Market Fairness
share
Fair machine learning research has been primarily concerned with classification tasks that result in discrimination. As machine learning algorithms are applied in new contexts, however, the harms or injustices that result are qualitatively different than those presently studied. Existing research at the level of metrics and definitions cannot measure these qualitatively different types of injustice. One example of this is the problem of market fairness and algorithmic collusion. Negative consequences of algorithmic collusion affect all consumers, not only particular members of a protected class. Drawing on this case study, I develop an ethical framework for fair machine learning research in new domains. This contribution ties the development of fairness metrics to specifically scoped normative principles. This enables fairness metrics to reflect different concerns from discrimination. I develop this framework and provide the philosophical rationale for its structure, ultimately applying it to the case of algorithmic collusion. I conclude with limitations of my proposal and discuss promising avenues of future research.
READ FULL TEXT