Informations générales
Intitulé de l'offre : PhD Student – Adversarial Attacks on Neural Networks [M/F] (H/F)
Référence : UMR8243-MELGOD-004
Nombre de Postes : 1
Lieu de travail : PARIS 13
Date de publication : jeudi 10 juillet 2025
Type de contrat : CDD Doctorant
Durée du contrat : 36 mois
Date de début de la thèse : 1 octobre 2025
Quotité de travail : Complet
Rémunération : 2200 gross monthly
Section(s) CN : 06 - Sciences de l'information : fondements de l'informatique, calculs, algorithmes, représentations, exploitations
Description du sujet de thèse
This thesis will focus on the study of adversarial attacks against neural networks, drawing on structural analogies between these models and block ciphers. It will specifically explore the adaptation of classical attack techniques from cryptanalysis — such as differential and linear attacks — to the context of artificial intelligence. The objective is twofold: to identify new types of attacks and to propose appropriate countermeasures to enhance the robustness of these models.
Contexte de travail
The team has expertise in cryptographic methods, algorithms, complexity theory, and neural networks, and includes among its permanent members notable researchers such as Simon Apers, Christina Boura, Geoffroy Couteau, Pierre Fraigniaud, Iordanis Kerenidis, Sophie Laplante, Frédéric Magniez, Claire Mathieu, Micuele Orrù, and Adrian Vladu.
Contraintes et risques
No risks or constraints identified.