Exploiting Edge Features for Transferable Adversarial Attacks in Distributed Machine Learning

Documenti analoghi

• Increasing the Confidence of Deep Neural Networks by Coverage Analysis
  di: Rossolini, Giulio, et al.
  Pubblicazione: (2021)
• Edge-Only Universal Adversarial Attacks in Distributed Learning
  di: Rossolini, Giulio, et al.
  Pubblicazione: (2024)
• Defending From Physically-Realizable Adversarial Attacks Through Internal Over-Activation Analysis
  di: Rossolini, Giulio, et al.
  Pubblicazione: (2022)
• Attention-Based Real-Time Defenses for Physical Adversarial Attacks in Vision Applications
  di: Rossolini, Giulio, et al.
  Pubblicazione: (2023)
• How Worst-Case Are Adversarial Attacks? Linking Adversarial and Perturbation Robustness
  di: Rossolini, Giulio
  Pubblicazione: (2026)