On Fri, May 19, 2000 at 04:28:22AM -0400, Dan Fabulich wrote:
> Makk Gingel suggested:
>
> > The hypothesis: It is possible, at least in principle, to construct a
> > machine which acquires all the skills and knowledge it requires to
> > function in the world via learning. Human provided domain-specific
> > knowledge, even of highly abstract or primitive domains, is
> > unnecessary. The question then is what is the minimal such machine,
> > and how do we go about building it.
>
> No. This is how you get a useful program given millions of years and
> plenty of trials.
That's entirely dependent upon the speed of your computers. It's quite
feasible for a simulated environment to outrun the real world by huge
factors.
> Alternately, this is how you get a useful program
> when you can get instant feedback as to how good your general
> "learning system" is, so finding the best one is just a matter of
> searching for it. (The latter is another way of saying the former.
> "Instant" is a point of view.)
This is the key point. Self-imroving systems have to have a way of
evaluating their success. The main challenge when applying one of the
self-development methods is setting the criteria for success.
Martin
(pet project: simulated two-legged walking robot connected to neural
network...)
-- +--------------------------------------------------------+ | Martin J. Ling Tel: +44 (0)20 8863 2948 | | martin@nodezero.org.uk Fax: +44 (0)20 8248 4025 | | http://www.nodezero.org.uk Mobile: +44 (0)7940 482675 | +--------------------------------------------------------+
This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:11:24 MDT