On 6/3/08, Jiri Jelinek <[EMAIL PROTECTED]> wrote:
>
> FYI, I recently run into some issues with my [under-development]
> formal language (which is being designed for my AGI-user
> communication) when trying to express statements like:
>
> "John said that if he knew yesterday what he knows today, he wouldn't
> do what he did back then."
Note that such problems have been addressed by classical AI theorists
a long time ago, though I'm not advocating that logic-based AI should
be "orthodox".
I have not investigated temporal logic in detail, so the exact
formalism I'll use is still open. Below I try to translate your
example using fluent calculus:
1. T(say(John,s1),past)
2. s1 = T(know(John,k1),past) -> ~T(will(do(a1)),past)
3. T(know(John,k1),today)
4. T(do(John,a1),past)
where: s1 is a sentence, k1 is a piece of knowledge, a1 is an action.
#3 states that John knows k1 today
#4 states that John has done action a1 in the past
Interestingly, there is an example in the book AIMA (2nd ed, p344):
"Lois believes today that Superman could fly yesterday"
T(Believes(Lois,"T(Flies(Superman),Yesterday)",Today).
YKY
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?member_id=8660244&id_secret=103754539-40ed26
Powered by Listbox: http://www.listbox.com
