September 2019
Volume 19, Issue 10
Open Access
Vision Sciences Society Annual Meeting Abstract  |   September 2019
Adversarial examples influence human visual perception
Author Affiliations & Notes
  • Gamaleldin F Elsayed
    Google Brain
  • Shreya Shankar
    Stanford University
  • Brian Cheung
    UC Berkeley
  • Nicolas Papernot
    Pennsylvania State University
  • Alexey Kurakin
    Google Brain
  • Ian Goodfellow
    Google Brain
  • Jascha Sohl-Dickstein
    Google Brain
Journal of Vision September 2019, Vol.19, 190c. doi:https://doi.org/10.1167/19.10.190c
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Gamaleldin F Elsayed, Shreya Shankar, Brian Cheung, Nicolas Papernot, Alexey Kurakin, Ian Goodfellow, Jascha Sohl-Dickstein; Adversarial examples influence human visual perception. Journal of Vision 2019;19(10):190c. https://doi.org/10.1167/19.10.190c.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Computer vision models are vulnerable to adversarial examples: small changes to images that cause models to make mistakes. Adversarial examples often transfer from one model to another, making it possible to attack models that an attacker has no access to. This raises the question of whether adversarial examples similarly transfer to humans. Clearly, humans are prone to many cognitive biases and optical illusions, but these generally do not resemble small perturbations, nor are they generated by optimization of a machine learning loss function. Thus, adversarial examples has been widely assumed – in the absence of experimental evidence – to not influence human perception. A rigorous investigation of the above question creates an opportunity both for machine learning and neuroscience. If we knew that the human brain could resist certain classes of adversarial examples, this would provide an existence proof for a similar mechanism in machine learning security. On the other hand, if we knew that the brain could be fooled by adversarial examples, this phenomenon could lead to a better understanding of brain function. Here, we investigate this question by leveraging three ideas from machine learning, neuroscience, and psychophysics[1]. First, we use black box adversarial example construction techniques to generate adversarial examples. Second, we adapt machine learning models to mimic the initial visual processing of humans. Third, we evaluate classification decisions of human observers in a time-limited setting to limit the brain’s utilization of recurrent and top-down processing pathways[2]. We find that adversarial examples that strongly transfer across computer vision models influence the classifications made by time-limited human observers. [1] A version of this work is accepted, but not yet presented, as a conference paper at NIPS 2018. [2] M. Potter et al. Detecting meaning in rsvp at 13 ms per picture. Attention, Perception, Psychophysics, 76(2):270–279, 2014.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×