thomblake comments on Compartmentalization as a passive phenomenon - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (71)
The second link isn't really about analogy-making as topic within AI, it's more about "analogy as flawed human thinking". (And Kaj's post reminds us precisely that given the role played by analogy in cognition, it may not fully deserve the bad rap Eliezer has given it.)
The first is partly about AI creativity (and also quite a bit about the flawed human thinking of AI researchers). It is the only one tagged "creativity"; and my reading of the Sequences has left me with an impression that the promise in the final sentence was left unfulfilled when I came to the end. I could rattle off a list of things I've learned from the Sequences, at various levels of understanding; they'd cover a variety of topics but creativity would be ranked quite low.
I mean, CopyCat comes up once in search results. If the topic of analogy within AI was discussed much here, I'd expect it to be referenced more often.
Remember, there is a long tradition here, especially for EY, to usually not refer to any scholarly research.