ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1711.08478
41
247

MagNet and "Efficient Defenses Against Adversarial Attacks" are Not Robust to Adversarial Examples

22 November 2017
Nicholas Carlini
D. Wagner
    AAML
ArXivPDFHTML
Abstract

MagNet and "Efficient Defenses..." were recently proposed as a defense to adversarial examples. We find that we can construct adversarial examples that defeat these defenses with only a slight increase in distortion.

View on arXiv
Comments on this paper