Communication Dans Un Congrès Année : 2025

Efficient Robust Conformal Prediction via Lipschitz-Bounded Networks

Résumé

Conformal Prediction (CP) has proven to be an effective post-hoc method for improving the trustworthiness of neural networks by providing prediction sets with finite-sample guarantees. However, under adversarial attacks, classical conformal guarantees do not hold anymore: this problem is addressed in the field of Robust Conformal Prediction. Several methods have been proposed to provide robust CP sets with guarantees under adversarial perturbations, but, for large scale problems, these sets are either too large or the methods are too computationally demanding to be deployed in real life scenarios. In this work, we propose a new method that leverages Lipschitz-bounded networks to precisely and efficiently estimate robust CP sets. When combined with a 1-Lipschitz robust network, we demonstrate that our lip-rcp method outperforms state-of-the-art results in both the size of the robust CP sets and computational efficiency in medium and large-scale scenarios such as ImageNet. Taking a different angle, we also study vanilla CP under attack, and derive new worst-case coverage bounds of vanilla CP sets, which are valid simultaneously for all adversarial attack levels. Our lip-rcp method makes this second approach as efficient as vanilla CP while also allowing robustness guarantees.

Fichier principal
Vignette du fichier
_ArXiv__Efficient_Robust_Conformal_Prediction_via_Lipschitz_Bounded_Networks.pdf (4.92 Mo) Télécharger le fichier
[ArXiv] Efficient Robust Conformal Prediction via Lipschitz-Bounded Networks.zip (4.42 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04936823 , version 1 (09-02-2025)
hal-04936823 , version 2 (04-06-2025)
hal-04936823 , version 3 (05-06-2025)

Licence

Identifiants

  • HAL Id : hal-04936823 , version 1

Citer

Thomas Massena, Léo Andéol, Thibaut Boissin, Franck Mamalet, Corentin Friedrich, et al.. Efficient Robust Conformal Prediction via Lipschitz-Bounded Networks. Forty-Second International Conference on Machine Learning (ICML 2025), Jul 2025, Vancouver, Canada. ⟨hal-04936823v1⟩
731 Consultations
748 Téléchargements

Partager

  • More