Robustness of deep learning classification to adversarial input on GPUs: asynchronous parallel accumulation is a source of vulnerability

Documenti analoghi

• Impacts of floating-point non-associativity on reproducibility for HPC and deep learning applications
  di: Shanmugavelu, Sanjif, et al.
  Pubblicazione: (2024)
• Aergia: Leveraging Heterogeneity in Federated Learning Systems
  di: Cox, Bart, et al.
  Pubblicazione: (2022)
• Roadmap for Edge AI: A Dagstuhl Perspective
  di: Ding, Aaron Yi, et al.
  Pubblicazione: (2021)
• Towards Optimal Heterogeneous Client Sampling in Multi-Model Federated Learning
  di: Zhang, Haoran, et al.
  Pubblicazione: (2025)
• Parameterizing Federated Continual Learning for Reproducible Research
  di: Cox, Bart, et al.
  Pubblicazione: (2024)