Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.11722 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911442601508864 |
|---|---|
| author | Bastian, Julien Leblanc, Benjamin Germain, Pascal Habrard, Amaury Largeron, Christine Metzler, Guillaume Morvant, Emilie Viallard, Paul |
| author_facet | Bastian, Julien Leblanc, Benjamin Germain, Pascal Habrard, Amaury Largeron, Christine Metzler, Guillaume Morvant, Emilie Viallard, Paul |
| contents | Classical PAC generalization bounds on the prediction risk of a classifier are insufficient to provide theoretical guarantees on fairness when the goal is to learn models balancing predictive risk and fairness constraints. We propose a PAC-Bayesian framework for deriving generalization bounds for fairness, covering both stochastic and deterministic classifiers. For stochastic classifiers, we derive a fairness bound using standard PAC-Bayes techniques. Whereas for deterministic classifiers, as usual PAC-Bayes arguments do not apply directly, we leverage a recent advance in PAC-Bayes to extend the fairness bound beyond the stochastic setting. Our framework has two advantages: (i) It applies to a broad class of fairness measures that can be expressed as a risk discrepancy, and (ii) it leads to a self-bounding algorithm in which the learning procedure directly optimizes a trade-off between generalization bounds on the prediction risk and on the fairness. We empirically evaluate our framework with three classical fairness measures, demonstrating not only its usefulness but also the tightness of our bounds. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_11722 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PAC-Bayesian Generalization Guarantees for Fairness on Stochastic and Deterministic Classifiers Bastian, Julien Leblanc, Benjamin Germain, Pascal Habrard, Amaury Largeron, Christine Metzler, Guillaume Morvant, Emilie Viallard, Paul Machine Learning Classical PAC generalization bounds on the prediction risk of a classifier are insufficient to provide theoretical guarantees on fairness when the goal is to learn models balancing predictive risk and fairness constraints. We propose a PAC-Bayesian framework for deriving generalization bounds for fairness, covering both stochastic and deterministic classifiers. For stochastic classifiers, we derive a fairness bound using standard PAC-Bayes techniques. Whereas for deterministic classifiers, as usual PAC-Bayes arguments do not apply directly, we leverage a recent advance in PAC-Bayes to extend the fairness bound beyond the stochastic setting. Our framework has two advantages: (i) It applies to a broad class of fairness measures that can be expressed as a risk discrepancy, and (ii) it leads to a self-bounding algorithm in which the learning procedure directly optimizes a trade-off between generalization bounds on the prediction risk and on the fairness. We empirically evaluate our framework with three classical fairness measures, demonstrating not only its usefulness but also the tightness of our bounds. |
| title | PAC-Bayesian Generalization Guarantees for Fairness on Stochastic and Deterministic Classifiers |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.11722 |