A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities

Documenti analoghi

• Scaling Laws from Sequential Feature Recovery: A Solvable Hierarchical Model
  di: Wortsman-Zurich, Arie, et al.
  Pubblicazione: (2026)
• Universality laws for Gaussian mixtures in generalized linear models
  di: Dandi, Yatin, et al.
  Pubblicazione: (2023)
• Asymptotics of feature learning in two-layer networks after one gradient-step
  di: Cui, Hugo, et al.
  Pubblicazione: (2024)
• Deep Learning of Compositional Targets with Hierarchical Spectral Methods
  di: Tabanelli, Hugo, et al.
  Pubblicazione: (2026)
• How Two-Layer Neural Networks Learn, One (Giant) Step at a Time
  di: Dandi, Yatin, et al.
  Pubblicazione: (2023)