BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Mathematical Theory of Robustness of Neural Networks
DTSTART:20230207T101500
DTEND:20230207T121500
DTSTAMP:20260916T032134Z
UID:2e71dccbcc680673d082a2b0bcef5f9b5b8822c73bf3ab3af19009c6
CATEGORIES:Conferences - Seminars
DESCRIPTION:Thomas Weinberger\nEDIC candidacy exam\nExam president: Prof.N
 icolas Flammarion\nThesis advisor: Prof. Rüdiger Urbanke\nCo-examiner: Pr
 of. Lenaic Chizat\n\nAbstract\ncoming soon\n\nBackground papers\nGradient 
 Methods Provably Converge to Non-Robust Networks\nA single gradient step f
 inds adversarial examples on random two-layers neural networks\nAdversaria
 lly Robust Generalization Requires More Data
LOCATION:
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR