Jürgen Schmidhuber(@SchmidhuberAI):This "recurrent depth" is essentially what's in Sec. 5.3 of the 2015 paper: On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models https://t.co/5FQEb7Cc3F. This paper went beyond the inefficient millisecond by millisecond planning of my 1990 neural world models, addressing planning and reasoning in abstract concept spaces. The 2015 control network C is a prompt engineer that learns to create a chain of thought: to speed up decision making, C learns to query its separate neural world model for abstract reasoning. The prompts and the answers are internal self-generated sequences of vectors that don't have to represent natural language.

2026.09.02 17:32

This "recurrent depth" is essentially what's in Sec. 5.3 of the 2015 paper: On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models This paper went beyond the inefficient millisecond by millisecond planning of my 1990 neural world models, addressing planning and reasoning in abstract concept spaces. The 2015 control network C is a prompt engineer that learns to create a chain of thought: to speed up decision making, C learns to query its separate neural world model for abstract reasoning. The prompts and the answers are internal self-generated sequences of vectors that don't have to represent natural language.

显示更多