On 25/06/07, Robert Wensman <[EMAIL PROTECTED]> wrote:
To put it bluntly: Isn't the concept of reinforcement learning (as typically
described) exactly the kind of simplistic AI models that AGI tries to
distance itself from?

It can be.  The sorts of problem which I've seen RL applied to were of
a fairly trivial nature, such as obstacle avoidance.  It seems that
situations where the internal architecture of the system has a
homogeneous structure usually don't scale well and this is often the
case in RL examples.

However I wouldn't throw RL out as a useful paradigm.  If you have a
look at Edelman's Darwin automata these are basically machines working
on a RL principle, but their internal architecture has a degree of
complexity more reminiscent of biological systems..

-----
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=e9e40a7e

Reply via email to