Aug 9, 2017 · 1 min read
Thanks for the link to the paper.
I’m agreement with the paper that analogical reasoning is used for understanding of fuzzy concepts in the real-world and that once concepts are identified, one can use rational approaches ( that are more rigorous forms of analogy).
What I am saying though is that our human circuitry to perform rational reasoning is learned and not intrinsic circuitry. So we use a parallel approximating cognitive mechanism to perform rigorous logic.
The reason why this is import, is that to build AGI we should build intuition machines that simulate rational machines. Classic AI has taken the opposite approach, rational machines that have never been able to simulate intuition machines.
