From: Peter Voss (peter@optimal.org)
Date: Mon Dec 30 2002 - 11:41:43 MST
Ben>Michael, I think this is a great direction to be thinking in, and a good
.5'th draft..
I agree. Thinking through AGI training/ teaching programs and 'IQ' tests is
very valuable.
Ben>I think that everything up thru section 13 in your document could be
done fairly straightforwardly by a genetic programming system with a
reasonable number of training examples and a reasonable amount of processing
power....
I strongly believe that during AGI proof-of-concept development, it is
crucial that we not only look at what the system accomplishes, but *how* it
does it - i.e. the design's construct validity.
Issues such as theoretical scalability and incremental (and one-shot)
learning need to be taken into account. I think it is also important for
early systems/tests to both deal with noisy environments and sense data far
in excess of what the system can handle.
One of the really difficult issues is designing these development programs
without embodying too much of the designer's specific AGI theory/
architecture.
Peter
PS. I've only looked at the first dozen or so lessons in Michael's paper.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT