Now showing items 1-2 of 2

    • Thumbnail

      Analysis of dominant classes in universal adversarial perturbations 

      Vadillo Jueguen, Jon; Santana Hermida, Roberto ORCID; Lozano Alonso, José Antonio (Elsevier, 2022-01)
      The reasons why Deep Neural Networks are susceptible to being fooled by adversarial examples remains an open discussion. Indeed, many different strategies can be employed to efficiently generate adversarial attacks, some ...
    • Thumbnail

      On the human evaluation of universal audio adversarial perturbations 

      Vadillo Jueguen, Jon; Santana Hermida, Roberto ORCID (Elsevier, 2022-01)
      [EN] Human-machine interaction is increasingly dependent on speech communication, mainly due to the remarkable performance of Machine Learning models in speech recognition tasks. However, these models can be fooled by ...