Provable Benefit of Sign Descent: A Minimal Model Under Heavy-Tailed Class Imbalance

Documenti analoghi

• Heavy-Tailed Class Imbalance and Why Adam Outperforms Gradient Descent on Language Models
  di: Kunstner, Frederik, et al.
  Pubblicazione: (2024)
• Adaptive Heavy-Tailed Stochastic Gradient Descent
  di: Gong, Bodu, et al.
  Pubblicazione: (2025)
• $(ε, u)$-Adaptive Regret Minimization in Heavy-Tailed Bandits
  di: Genalti, Gianmarco, et al.
  Pubblicazione: (2023)
• Heavy-Tailed Class-Conditional Priors for Long-Tailed Generative Modeling
  di: Bouayed, Aymene Mohammed, et al.
  Pubblicazione: (2025)
• Provable Benefits of In-Tool Learning for Large Language Models
  di: Houliston, Sam, et al.
  Pubblicazione: (2025)