Notice détaillée
Titre
TML
Formal Name (French)
Laboratoire de Théorie en Apprentissage Automatique
Formal Name (English)
Theory of Machine Learning Laboratory
Lab Manager
Flammarion, Nicolas
Group ID
U13719
Auteurs affilié
Andriushchenko, Maksym
Bachmann Ona, Jennifer
Boursier, Etienne
Croce, Francesco
D'Angelo, Francesco
Flammarion, Nicolas
Papazov, Hristo Georgiev
Pesme, Scott
Pillaud-Vivien, Loucas
Varre, Aditya Vardhan
Vladarean, Maria-Luiza
Yüce, Gizem
Yüksel, Oguz Kaan
Bachmann Ona, Jennifer
Boursier, Etienne
Croce, Francesco
D'Angelo, Francesco
Flammarion, Nicolas
Papazov, Hristo Georgiev
Pesme, Scott
Pillaud-Vivien, Loucas
Varre, Aditya Vardhan
Vladarean, Maria-Luiza
Yüce, Gizem
Yüksel, Oguz Kaan
Institut
IC
Faculté
IINFCOM
Note
Members of the TML unit
Lien extérieur
https://www.epfl.ch/labs/tml/
Publications
Escaping from saddle points on Riemannian manifolds
From averaging to acceleration, there is only a step-size
On Convergence-Diagnostic based Step Sizes for Stochastic Gradient Descent
On the effectiveness of adversarial training against common corruptions
RobustBench: a standardized adversarial robustness benchmark
Saddle-to-Saddle Dynamics in Diagonal Linear Networks
Sparse-RS: A Versatile Framework for Query-Efficient Sparse Black-Box Adversarial Attacks
Sparse-RS: a versatile framework for query-efficient sparse black-box adversarial attacks
Trace norm regularization for multi-task learning with scarce data
Understanding and Improving Fast Adversarial Training
Voir toutes les publications (38)
From averaging to acceleration, there is only a step-size
On Convergence-Diagnostic based Step Sizes for Stochastic Gradient Descent
On the effectiveness of adversarial training against common corruptions
RobustBench: a standardized adversarial robustness benchmark
Saddle-to-Saddle Dynamics in Diagonal Linear Networks
Sparse-RS: A Versatile Framework for Query-Efficient Sparse Black-Box Adversarial Attacks
Sparse-RS: a versatile framework for query-efficient sparse black-box adversarial attacks
Trace norm regularization for multi-task learning with scarce data
Understanding and Improving Fast Adversarial Training
Voir toutes les publications (38)
Le document apparaît dans
Authorities > Lab