Theoretical evidence for adversarial robustness
          through randomization

         Rafael Pinot1,2 Laurent Meunier1,3 Alexandre Araujo1,4
      Hisashi Kashima5,6 Florian Yger1 Cédric Gouy-Pailler2 Jamal Atif1
    1
     Université Paris-Dauphine, PSL Research University, CNRS, LAMSADE, Paris, France
     2
      Institut LIST, CEA, Université Paris-Saclay 3 Facebook AI Research, Paris, France
  4
   Wavestone, Paris, France 5 Kyoto University, Kyoto, Japan 6 RIKEN Center for AIP, Japan

...