Everything is Relative: Understanding Fairness with Optimal Transport
This work addresses the challenge for practitioners in selecting and applying fairness definitions by providing a more interpretable and quantifiable method to analyze bias in AI systems, though it is incremental in building on existing optimal transport techniques.
The authors tackled the problem of interpreting and quantifying fairness in automated decision-making systems by proposing an optimal transport-based approach that compares outcomes to reveal bias structure, enabling detection of unfairness where other metrics fail and exploration of recourse opportunities.
To study discrimination in automated decision-making systems, scholars have proposed several definitions of fairness, each expressing a different fair ideal. These definitions require practitioners to make complex decisions regarding which notion to employ and are often difficult to use in practice since they make a binary judgement a system is fair or unfair instead of explaining the structure of the detected unfairness. We present an optimal transport-based approach to fairness that offers an interpretable and quantifiable exploration of bias and its structure by comparing a pair of outcomes to one another. In this work, we use the optimal transport map to examine individual, subgroup, and group fairness. Our framework is able to recover well known examples of algorithmic discrimination, detect unfairness when other metrics fail, and explore recourse opportunities.