From: Ben Goertzel (ben@goertzel.org)
Date: Thu Sep 08 2005 - 16:11:58 MDT
Michael Wilson wrote:
> > No, the attack I am making is that there are an array of subtle faults
> > that the Bayesian approach cannot get out of
>
> Which you have yet to describe, or rather have tried and failed to
> describe one.
Well, my issue with a purely probabilistic approach is that it doesn't seem
to give a computationally tractable approach to the formation of complex
concepts or predicates.
Once I have a hypothesis, probabilistic inference (including Bayes rule) is
useful for estimating its truth value.
But what about hypothesis formation? The space of possible hypotheses is
very large. Most of AI is devoted to search techniques for searching
hypothesis space, and these techniques suck in an AGI context.
An AGI needs a concept/hypothesis formation approach that plays well with
probabilistic inference, but I have never seen an adequate
concept/hypothesis formation method that's based solely on probabilistic
concepts. Do you have one that you'd care to discuss?
-- Ben
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:52 MDT