Communication Dans Un Congrès Année : 2025

Robust ML Auditing using Prior Knowledge

Résumé

Among the many technical challenges to enforcing AI regulations, one crucial yet underexplored problem is the risk of audit manipulation. This manipulation occurs when a platform deliberately alters its answers to a regulator to pass an audit without modifying its answers to other users. In this paper, we introduce a novel approach to manipulation-proof auditing by taking into account the auditor's prior knowledge of the task solved by the platform. We first demonstrate that regulators must not rely on public priors (e.g. a public dataset), as platforms could easily fool the auditor in such cases. We then formally establish the conditions under which an auditor can prevent audit manipulations using prior knowledge about the ground truth. Finally, our experiments with two standard datasets illustrate the maximum level of unfairness a platform can hide before being detected as malicious. Our formalization and generalization of manipulation-proof auditing with a prior opens up new research directions for more robust fairness audits.

Fichier principal
Vignette du fichier
7115_Robust_ML_Auditing_using_.pdf (578.29 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05268400 , version 1 (19-09-2025)

Licence

Identifiants

Citer

Jade Garcia Bourrée, Augustin Godinot, Sayan Biswas, Anne-Marie Kermarrec, Erwan Le Merrer, et al.. Robust ML Auditing using Prior Knowledge. ICML 2025 - 42nd International Conference on Machine Learning, Jul 2025, Vancouver, Canada. pp.1-17, ⟨10.48550/arXiv.2505.04796⟩. ⟨hal-05268400⟩
694 Consultations
203 Téléchargements

Altmetric

Partager

  • More