Stanley, 
Thagard's work on coherence is pretty interesting.  He combines it with 
Damasio's work to form a pretty cogent theory. Similar to yours...
http://cogsci.uwaterloo.ca/Articles/Pages/how-to-decide.html

http://cogsci.uwaterloo.ca/Articles/Pages/Inference.Plan.html

http://cogsci.uwaterloo.ca/Articles/Pages/Emot.Decis.html

http://cogsci.uwaterloo.ca/Articles/Pages/Empathy.html
Cheers.
~PM

Date: Wed, 15 May 2013 07:23:49 -0600
From: [email protected]
To: [email protected]
CC: [email protected]
Subject: Re: [agi] Goal Selection


  
    
  
  
    On 05/14/2013 03:56 PM, Piaget Modeler
      wrote:

    
    
      
      

        My intuition tells me that
            there should be another, higher level of goal selection.
            (Or, perhaps goal
        filtering, not exactly
            sure). Something that operates above action selection that
            takes into account 
        all the possible goals the system could
            have and ensures that the most important at the current 
        moment are part of the
            agenda.  
         
        Your thoughts?
        

        
        ~PM
        

      
    
    Stay with your intuition.  consider - 

     - take into account all the possible goals the system could have

        call those "all the opportunities" 

    

     - "ensure that the most important at the moment"

        call that the best opportunity

    

     - "are part of the agenda" 

        that's the "arbitrator" who is looking at the current activity
    along with the proposed best opportunities.  Either it's "best" to
    stay with the current opportunity or go with the "filtered up" best
    opportunity.   

    

    A scheme like this requires that every opportunity be "ranked" and
    promoted at the proper time.  The promotion is the easy part - when
    conditions are met, do the promotion.  The ranking is the difficult
    part - hard but not impossible.  

    

    Arbitration can be simple, look at rank.  More sophisticated
    Arbitrator might do an analysis of fit with the long term strategy.
    

    

    I think this kind of design has promise as it is easy to understand
    and allows for instrumentation as it would be simple to log the
    "current" opportunity and see how it changes over time. 

    

    This is a quick sketch of my current design. 

    Stan

    

    

  


  
    
      
      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-f452e424
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com

Reply via email to