Ricardo Barreira wrote:
But NP-hard is still very, very trivial; there's very little that
falls into
that category. Basically there are four important categories of problem:
1) NP-hard
2) EXPTIME-hard
3) Incomputable
4) Ill-posed problem
Intelligence is primarily in category 4 (which is Richard's point, if I
understand him correctly).
Which doesn't mean we can't prove that it's also in other categories
(even if the problem isn't completely defined); that was Eliezer's
assertion, which seemed pretty fair to me and which Richard tried to
disprove with an unhelpful argument.
Ricardo,
No, you misunderstand me. You say "Which doesn't mean we can't prove
that it's also in other categories....", however, what I pointed out was
that you cannot define the 'it' to which you refer in this sentence.
If you cannot say what 'it' is (what intelligence is) then how can you
make any statements about whether it is incomputable, (EXPTIME-hard or
NP-hard? The only type of concept to which these three things can
possibly be applied is a concept that has a well defined meaning.
If you think my argument is not valid, you will have to explain to me
how something can be both undefined and yet at the same time subject to
a proof of how computationally difficult it is.
I'm not sure why you would think it an "unhelpful argument." Isn't it a
clear case of semantic incoherence?
Richard Loosemore
-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
