Adversarial Examples in Simpler Settings

In this thesis we explore adversarial examples for simple model families and simple data distributions, focusing in particular on linear and kernel classifiers. On the theoretical front we find evidence that natural accuracy and robust accuracy are more likely than not to be misaligned. We conclude...

Descrizione completa

Dettagli Bibliografici
Autore principale: Wang, Tony T.
Altri autori: Wornell, Gregory W.
Natura: Tesi
Pubblicazione: Massachusetts Institute of Technology 2022
Accesso online:https://hdl.handle.net/1721.1/139041