1 paper
François Menet, Paul Berthier, José M. Fernandez +1
Deep learning models are vulnerable to adversarial examples which are input samples modified in order to maximize the error on the system. We introduce Spartan Networks, resistant…