The effects of Hessian eigenvalue spectral density type on the applicability of Hessian analysis to generalization capability assessment of neural networks

Documenti analoghi

• Investigating generalization capabilities of neural networks by means of loss landscapes and Hessian analysis
  di: Gabdullin, Nikita
  Pubblicazione: (2024)
• Exploring possible vector systems for faster training of neural networks with preconfigured latent spaces
  di: Gabdullin, Nikita
  Pubblicazione: (2025)
• Using predefined vector systems to speed up neural network multimillion class classification
  di: Gabdullin, Nikita, et al.
  Pubblicazione: (2026)
• Using predefined vector systems as latent space configuration for neural network supervised training on data with arbitrarily large number of classes
  di: Gabdullin, Nikita
  Pubblicazione: (2025)
• Hessian Surgery: Class-Targeted Post-Hoc Rebalancing via Hessian Spike Perturbation
  di: Vigna, Hugo, et al.
  Pubblicazione: (2026)