no code implementations • 9 Oct 2023 • Teresa Scantamburlo, Joachim Baumann, Christoph Heitz
We clarify the distinction between the concepts of prediction and decision and show the different ways in which these two elements influence the final fairness properties of a prediction-based decision system.
1 code implementation • 19 Oct 2022 • Joachim Baumann, Christoph Heitz
In this paper, we present a step-by-step procedure integrating three elements: (a) a framework for the moral assessment of what fairness means in a given context, based on the recently proposed general principle of "Fair equality of chances" (FEC) (b) a mapping of the assessment's results to established statistical group fairness criteria, and (c) a method for integrating the thus-defined fairness into optimal decision making.
1 code implementation • 6 Jun 2022 • Joachim Baumann, Corinna Hertweck, Michele Loi, Christoph Heitz
Group fairness metrics are an established way of assessing the fairness of prediction-based decision-making systems.
2 code implementations • 6 Jun 2022 • Corinna Hertweck, Joachim Baumann, Michele Loi, Eleonora Viganò, Christoph Heitz
This allows us to derive a fairness score that we then compare to the decision maker's utility.
1 code implementation • 5 Jun 2022 • Joachim Baumann, Anikó Hannák, Christoph Heitz
We show that group-specific threshold rules are optimal for PPV parity and FOR parity, similar to well-known results for other group fairness criteria.
no code implementations • 11 May 2022 • Michele Loi, Christoph Heitz
For our paper, we equate fairness with (non-)discrimination, which is a legitimate understanding in the discussion about group fairness.
no code implementations • 9 Sep 2021 • Corinna Hertweck, Christoph Heitz
While the field of algorithmic fairness has brought forth many ways to measure and improve the fairness of machine learning models, these findings are still not widely used in practice.
no code implementations • 4 Nov 2020 • Corinna Hertweck, Christoph Heitz, Michele Loi
This means that the question of whether independence should be used or not cannot be satisfactorily answered by only considering the justness of differences in the predictive features.