Sequential Conditional Transport on Probabilistic Graphs for Interpretable Counterfactual Fairness
Abstract
In this paper, we link two existing approaches to derive counterfactuals: adaptations based on a causal graph, and optimal transport. We extend "Knothe's rearrangement" and "triangular transport" to probabilistic graphical models, and use this counterfactual approach, referred to as sequential transport, to discuss fairness at the individual level. After establishing the theoretical foundations of the proposed method, we demonstrate its application through numerical experiments on both synthetic and real datasets.
Cite
Text
Machado et al. "Sequential Conditional Transport on Probabilistic Graphs for Interpretable Counterfactual Fairness." AAAI Conference on Artificial Intelligence, 2025. doi:10.1609/AAAI.V39I18.34131Markdown
[Machado et al. "Sequential Conditional Transport on Probabilistic Graphs for Interpretable Counterfactual Fairness." AAAI Conference on Artificial Intelligence, 2025.](https://mlanthology.org/aaai/2025/machado2025aaai-sequential/) doi:10.1609/AAAI.V39I18.34131BibTeX
@inproceedings{machado2025aaai-sequential,
title = {{Sequential Conditional Transport on Probabilistic Graphs for Interpretable Counterfactual Fairness}},
author = {Machado, Agathe Fernandes and Charpentier, Arthur and Gallic, Ewen},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2025},
pages = {19358-19366},
doi = {10.1609/AAAI.V39I18.34131},
url = {https://mlanthology.org/aaai/2025/machado2025aaai-sequential/}
}