Ellie Pavlick tells us "we decided that meaning involves concepts in some way... If you use the word 'apple' to mean apple, you need the concept of an apple. That has to be a thing, whether or not you use the word to refer to it." And my opinion is, this is a category error. So it was interesting to see them actually find evidence of such a concept - "we found a small place in the model where it basically boils that connection down into one little vector... It's like this systematic 'retrieve-capital-city' vector." Is that what it means to, say, 'know' what an apple is? If so, though, then the knowledge and the thing knowing are one and the same thing - there's no 'concept' over and above the 'conceiver'.
Today: 1 Total: 114 [Direct link] [Share]
