Skip to Main Content Area
  • CPS-VO
    • Contact Support
  • Browse
    • Calendar
    • Announcements
    • Repositories
    • Groups
  • Search
    • Search for Content
    • Search for a Group
    • Search for People
    • Search for a Project
    • Tagcloud
      
 
Not a member?
Click here to register!
Forgot username or password?
 
Home
National Science Foundation

Cyber-Physical Systems Virtual Organization

Read-only archive of site from September 29, 2023.

CPS-VO

black-box settings

biblio

Visible to the public GeoDA: A Geometric Framework for Black-Box Adversarial Attacks

Submitted by aekwall on Tue, 03/09/2021 - 12:05pm
  • image classification
  • Robustness
  • Resiliency
  • query processing
  • queries
  • pubcrawl
  • Perturbation methods
  • pattern classification
  • optimisation
  • Neural networks
  • natural image classifiers
  • minimal perturbation
  • Metrics
  • Measurement
  • mean curvature
  • Iterative methods
  • adversarial examples
  • geometric framework
  • gaussian distribution
  • estimation
  • effective iterative algorithm
  • Deep Networks
  • decision boundary
  • data samples
  • Covariance matrices
  • composability
  • carefully perturbed images
  • black-box settings
  • black-box perturbations
  • black-box attack algorithm
  • black-box adversarial attacks
  • black box encryption

Terms of Use  |  ©2023. CPS-VO