Theoretical Analysis of Relative Errors in Gradient Computations for Adversarial Attacks with CE Loss

Documenti analoghi

• RCR-AF: Enhancing Model Generalization via Rademacher Complexity Reduction Activation Function
  di: Yu, Yunrui, et al.
  Pubblicazione: (2025)
• Rethinking Gradient-based Adversarial Attacks on Point Cloud Classification
  di: Chen, Jun, et al.
  Pubblicazione: (2025)
• Adversarial Attacks on Medical Hyperspectral Imaging Exploiting Spectral-Spatial Dependencies and Multiscale Features
  di: Gu, Yunrui, et al.
  Pubblicazione: (2026)
• AEMIM: Adversarial Examples Meet Masked Image Modeling
  di: Xiang, Wenzhao, et al.
  Pubblicazione: (2024)
• Human-Imperceptible Physical Adversarial Attack for NIR Face Recognition Models
  di: Xie, Songyan, et al.
  Pubblicazione: (2025)