Bob M:  Minsky
says that one of the key things which an intelligent system ought to
be able to do is reason by analogy.

 "His thoughts tumbled in his head, making and breaking alliances
like underpants in a dryer without Cling Free."

I agree - analogy is central to AGI, hugely important. Ben doesn't specify how he draws analogies or give examples. And Pei's example:
robin --> bird> ==> <robin --> animal> %0.9%
<robin --> flyer> <=> <robin --> animal> %0.9%
<robin --> bird> ==> <robin --> flyer>
--------------------
<<robin --> bird> ==> <robin --> flyer>> %0.81;0.59%
seems like logic/ classification to me, rather than analogy.

Analogy and metaphor do seem to me to work by "drawing" - making connections by use of sensory graphics. and images. Connections which often cut across the senses - your "lost moments" could also (though not so aptly) be compared to "fading echoes" as well as "tears in rain."

I can't see how the brain's capacity, say, to see an "idea" as
*the key to a problem
*the solution (think liquid)
*a breakthrough
*a light bulb coming on
*hitting the jackpot

or per your example to see "thoughts associating" as

*like underpants tumbling in a dryer
*making and breaking alliances (a separate, sociopolitical metaphor)

can be other than based on graphics and images, (although it is often done in complex ways).

Does anyone have alternative mechanisms for analogy?

(I think this area is worth a lot of discussion - it's so important to AGI).


-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936

Reply via email to