When narrower is better: the narrow width limit of Bayesian parallel branching neural networks

Documenti analoghi

• Connecting NTK and NNGP: A Unified Theoretical Framework for Wide Neural Network Learning Dynamics
  di: Avidan, Yehonatan, et al.
  Pubblicazione: (2023)
• Memorization and Knowledge Injection in Gated LLMs
  di: Pan, Xu, et al.
  Pubblicazione: (2025)
• Coding schemes in neural networks learning classification tasks
  di: van Meegen, Alexander, et al.
  Pubblicazione: (2024)
• Simmering: Sufficient is better than optimal for training neural networks
  di: Babayan, Irina, et al.
  Pubblicazione: (2024)
• Order parameters and phase transitions of continual learning in deep neural networks
  di: Shan, Haozhe, et al.
  Pubblicazione: (2024)