Ben:Note, this is not a way of building an AGI, it's a way of modeling and 
understanding a mind

Would you care to expand? I accept that various current AGI projects are useful 
“models of the mind” -  vastly different from the real thing, but useful as 
exploratory forays for an eventual systemic science of the mind – useful for 
science. As you know, I wouldn’t accept that any AFAIK have anything to do with 
AGI – are useful for a real AGI technology.


From: Ben Goertzel 
Sent: Thursday, November 01, 2012 6:40 AM
To: AGI 
Subject: Re: [agi] The Fundamental Misunderstanding in AGI [was Superficiality]


The Fisher-Rao metric measures the distance between two probability 
distributions 

An individual memory state M corresponds to a certain distribution over memory 
states -- this is the distribution indicating how active each other memory 
state is, when you activate M

Note, this is not a way of building an AGI, it's a way of modeling and 
understanding a mind...

ben


On Wed, Oct 31, 2012 at 6:27 PM, YKY (Yan King Yin, 甄景贤) 
<[email protected]> wrote:

  Hi Ben and others,

  I have some questions about your paper "3 hypotheses about the geometry of 
mind": 

  1.  How do you acquire the probability distributions over memory states, 
initially?

  2.  Does the Fisher-Rao metric help in calculating the distributions?  Or it 
just provides the distance between 2 distributions?

  3.  How is the distance between 2 distributions related to the distance 
between 2 memory states?  it seems that we'd be interested in the latter.

  I've been skimming the beginning of the book by Amari and Nagaoka, but I 
don't see how the theory is applied...

  Thanks in advance =)
  KY
        AGI | Archives  | Modify Your Subscription  





-- 
Ben Goertzel, PhD
http://goertzel.org

"My humanity is a constant self-overcoming" -- Friedrich Nietzsche


      AGI | Archives  | Modify Your Subscription   



-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968
Powered by Listbox: http://www.listbox.com

Reply via email to