From Lazy to Rich: Exact Learning Dynamics in Deep Linear Networks

Documenti analoghi

• A Theory of How Pretraining Shapes Inductive Bias in Fine-Tuning
  di: Anguita, Nicolas, et al.
  Pubblicazione: (2026)
• Get rich quick: exact solutions reveal how unbalanced initializations promote rapid feature learning
  di: Kunin, Daniel, et al.
  Pubblicazione: (2024)
• Optimal Representation Size: High-Dimensional Analysis of Pretraining and Linear Probing
  di: Njaradi, Valentina, et al.
  Pubblicazione: (2026)
• Position: Solve Layerwise Linear Models First to Understand Neural Dynamical Phenomena (Neural Collapse, Emergence, Lazy/Rich Regime, and Grokking)
  di: Nam, Yoonsoo, et al.
  Pubblicazione: (2025)
• Flexible task abstractions emerge in linear networks with fast and bounded units
  di: Sandbrink, Kai, et al.
  Pubblicazione: (2024)