Non-Equilibrium Stochastic Dynamics as a Unified Framework for Insight and Repetitive Learning: A Kramers Escape Approach to Continual Learning

Documenti analoghi

• When can in-context learning generalize out of task distribution?
  di: Goddard, Chase, et al.
  Pubblicazione: (2025)
• Drift-Diffusion Matching: Embedding dynamics in latent manifolds of asymmetric neural networks
  di: Nartallo-Kaluarachchi, Ramón, et al.
  Pubblicazione: (2026)
• A statistical physics framework for optimal learning
  di: Mignacco, Francesca, et al.
  Pubblicazione: (2025)
• Data coarse graining can improve model performance
  di: Nguyen, Alex, et al.
  Pubblicazione: (2025)
• Introduction to dynamical mean-field theory of randomly connected neural networks with bidirectionally correlated couplings
  di: Zou, Wenxuan, et al.
  Pubblicazione: (2023)