melinaverger / madd

Model Absolute Density Distance (MADD)

This repository contains all the source code of the experiments that were done in the papers of the References section.

Note

A Python package has been developped here to evaluate and mitigate algorithmic unfairness with the MADD easily.

MADD metric and MADD post-processing

You can see more instructions in the READMEs from initial_code folder and in RKDE_ECMLPKK2023 > post_processing folder (on the one hand for evaluating with the MADD and on the other hand for mitigating based on the MADD).

References

The links of the published papers are here.

[under review] Mélina Verger, François Bouchet, Sébastien Lallé, Vanda Luengo. Discriminations intersectionnelles : approfondir l’évaluation de l’équité algorithmique pour la prédiction de la réussite à des cours en ligne. Revue STICEF. Edition numéro spécial «Sélection de la conférence EIAH 2023».

[under revision] Mélina Verger, Chunyang Fan, Sébastien Lallé, François Bouchet, Vanda Luengo. Evaluating and Mitigating Algorithmic Unfairness with the MADD Metric: An Optimal Computation Applied to Predictive Student Models. Journal of Educational Data Mining. Special edition from EDM 2023 conference.

Mélina Verger, Chunyang Fan, Sébastien Lallé, François Bouchet, Vanda Luengo. A Fair Post-Processing Method based on the MADD Metric for Predictive Student Models. 1st International Tutorial and Workshop on Responsible Knowledge Discovery in Education (RKDE 2023) at ECML PKDD 2023, September 2023, Turino, Italy.

Mélina Verger, Sébastien Lallé, François Bouchet, Vanda Luengo. Is Your Model "MADD"? A Novel Metric to Evaluate Algorithmic Fairness for Predictive Student Models. Sixteenth International Conference on Educational Data Mining (EDM 2023), July 2023, Bangalore, India.

Mélina Verger, François Bouchet, Sébastien Lallé, Vanda Luengo. Caractérisation et mesure des comportements discriminants des modèles prédictifs. 11ème Conférence sur les Environnements Informatiques pour l'Apprentissage Humain (EIAH 2023), June 2023, Brest, France.