187 citations · 721 across the 60 of their papers we have counts for
1 paper · 1 filter
Gerda Bortsova, Cristina González-Gonzalo, Suzanne C. Wetstein +9
Adversarial attacks are considered a potentially serious security threat for machine learning systems. Medical image analysis (MedIA) systems have recently been argued to be vulner…