Natural Abstraction Hypothesis

Creator
Creator
Seonglae ChoSeonglae Cho
Created
Created
2024 Oct 24 23:7
Editor
Edited
Edited
2024 Oct 24 23:15
Refs
Refs

NAH

AI가 학습하는 효율적인 추상화는 환경 자체의 특성을 반영
  • Abstractability - The physical world can be abstracted, and it can be summarized with information of a much lower dimension than the overall complexity of the system
  • Human-Compatibility - Low-dimensional abstraction aligns with the abstractions humans use
  • Convergence - Various cognitive structures are likely to use similar abstractions
 
 
 
 

World model Interpretability with
AI Internal Interface

If the way AI interacts with various modules through internal interfaces is consistently formed, the possibility increases that humans can understand the format of these interfaces and interpret the entire world model at once.
 
 

Recommendations