David Butler <[EMAIL PROTECTED]> wrote:Would two AGI's with the same initial 
learning program, same hardware in a controlled environment (same access to a 
specific learning base-something like an encyclopedia) learn at different rates 
and excel in different tasks? 

How would an AGI choose which things to learn first if given enough  data so 
that it would have to make a choice? If two AGI's (again-same  hardware, 
learning programs and controlled environment) were given  the same data would 
they make different choices?
Yes, any two exact copies of AGI's would learn at different rates, and learn 
different things, its one of the cores and basic easy things that can be 
programmed into an AGI, *with ease*.

To make a decision ever, an AGI takes all pertinent information, calculates it 
over its choices, and makes the choice with highest value, when any choice has 
approx the same value, it needs a tie-breaker.  If the machines just choose the 
First choice every time, then they will always act the same given the same 
input.  But a simple random number generator can tell the AGI, ok now I want to 
read encycolpedia A or Z, the first may choose A, the second Z, and they 
diverge from there.

This simple concept is very useful for making an AGI explore an alternative 
choice to the one they are given, perhaps allowing them to choose a seperate 
path to a goal that provides them with unique or new information, or providing 
a creative answer to a problem.

Im not sure what game it is from, but a VR AI program had a problem of 
attacking someone as they walked outside the house, and normally it was 
expected to go out the door and attack.  On one iteration though, to dove and 
rolled thru the window and attacked instead.  This was unexpected and creative 
solution to the problem at hand.

James Ratcliff


_______________________________________
James Ratcliff - http://falazar.com
Looking for something...
       
---------------------------------
Be a better friend, newshound, and know-it-all with Yahoo! Mobile.  Try it now.

-----
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=8660244&id_secret=88965806-34e98c

Reply via email to