From: Ben Goertzel (ben@goertzel.org)
Date: Wed May 02 2001 - 04:31:31 MDT
Logic-based AI and self-organizing-emergence based-AI both have their
strengths and weaknesses. To ignore the insights obtained by either
approach is a mistake. The key is to figure out how they work together.
When you really get into it, the most abstract and complex logic-based tasks
come out to look VERY similar to the lowest-level perceptual processes....
The common substrate of processes underlying the two is more significant
than the differences, though the differences are real too.
For instance, in the Webmind AI Engine, the hardest part of higher-order
inference is creating compound relations (boolean conjunctions of elementary
relations). These are then evaluated and interrelated using inference
rules, but guessing which compound relations to form in a given context is a
hard problem not solved by logic, rather solved by evolution-like methods.
Yet, this problem is very similar to recognizing patterns in a perceptual
stream -- where one is finding compound relations of elementary percepts
rather than elementary cognitive relations
ben
> > I was referring to the time when I read Elaine Rich,
> > Hofstadter and Minsky, and all kinds of weird AI journals, and believed
> > their approach was not sterile. This was about 15 years ago. Then I got
> > a bad case of cellular automata, complexity, resulting in processing of
> > lots of dead tree labeled with Fredkin, Toffioli, Holland,
> Koza, Kauffman,
> > Wolfram, and their illustrous ilk. I still haven't recovered yet.
> > I'm looking for other infections material, but so far haven't found
> > much. Maybe I've become immune, that would be a pity.
>
> The GOFAI approach based on pure logic has been pretty much
> proven sterile,
> although there have been a few successes, such as ("statistical") machine
> learning, constraint logic programming.
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:07:26 MST